Modelo de classificação via Random Forest
Uma aplicação com dados da plataforma IMDB - Internet Movie DataBases
Ewerson C. Pimenta
pimentaeu@yahoo.comMarcos Antônio E. de Oliveira
marcoseuzbio@gmail.comRio de Janeiro, 25 de outubro de 2018
ptm = proc.time()1 Introdução
Frequentemente utilizamos serviços de classificação, seja com o intuito de recomendar/sugerir algo a um usuário final, como Netflix, Mercado Livre, OLX dentre outros, ou seja para determinar em testes de diagnóstico de doenças se um paciente deve ser classificado como doente, não doente ou suspeito. Além de outros exemplos. O conjunto de dados explanados a seguir foram extraídos da plataforma Kaggle e são referentes ao catálogo de filmes da plataforma IMDB, o intuito aqui é criar um sistema de classificação de estrelas do catálogo de avaliação de filmes do IMDB.
Esse projeto irá explorar o conjunto de dados a fim de gerar insights, e aplicar o algoritmo de Random Forest - RF (baseado na clusterização de Breiman) para classificação de filmes nas seguintes faixas/categorias de estrelas (Rating):
- Categoria ‘Baixo índice de aprovação (low score)’: [0, 6.5) ★
- Categoria ‘Alto índice de aprovação (high score)’: [6.5, 10) ★
Para este, se faz necessário um largo conjunto de dados que, por sua vez, deve ser bem estruturado e consistente. Portanto, é fundamental manipular, limpar e remover informações faltantes dos dados para a implementação do modelo de Machine Learning do Random Forest que será treinado com orientação a esses dados.
Além disso, esse trabalho irá levantar quais os fatores mais importantes para que um filme tenha uma alta classificação (categoria de faixas de estrelas), via modelagem de Random Forest, de tal forma que essas mesmas variáveis/fatores destacados como importantes no algoritmo do RF possam ser aproveitados posteriormente em análises futuras de classificação. Resultados e script foram gerados utilizando recursos e linguagens em \(R\), \(\LaTeX\) e \(Markdown\).
Pacotes e Leitura
Alguns pacotes R utilizados nessa rotina
library(tidyverse) # Manipulação de banco de dados e análise exploratória
library(highcharter) # visualização Gráfica
library(DT) # Construção de Tabelas
library(knitr) # Opções Rmarkdown (inclusão de tabelas, imagens, etc.)
library(kableExtra) # Construção de Tabelas
library(rpart.plot) # Recursive Partitioning and Regression Trees
library(randomForest) # Modelo Random Forest
library(corrplot) # Matriz de correlação - Visualização gráfica
library(wordcloud2) # Numvem de palavras (Word Cloud)
library(caret) # Matriz de confusão (Confusion Matrix)
library(rpart)
#devtools::install_github("ropenscilabs/icon")
library(icon)RECOMEND.METADATA = readxl::read_xlsx(path = ".../DB/IMDB.xlsx", sheet = 1)
db = RECOMEND.METADATA+ + =
2 Banco de Dados
O conjunto de dados utilizado foi o de título IMDB 5000 extraído da plataforma kaggle. As informações contidas no banco foram catalogadas de filmes publicados ao longo de 100 anos em 66 países (entre 1916 e 2016) da plataforma IMDB - Internet Movie DataBases, o Arquivo original contém 5044 filmes (observações) e 28 variáveis descritas a seguir.
Os dados em questão são públicos e disponíveis para download clicando AQUI.
3 Análise Exploratória dos dados
## Warning in instance$preRenderHook(instance): It seems your data is too
## big for client-side DataTables. You may consider server-side processing:
## https://rstudio.github.io/DT/server.html3.1 Estrutura dos dados
## O banco de dados possui 5043 observações e 28 variáveisExcluindo algumas variáveis que não serão utilizadas.
- As variáveis color, director_name, language, country, content_rating, em especial, foram removidas uma vez que estas não se mostraram importantes no algoritmo de importância do Random Forest de Breiman.
- Já as variáveis director_facebook_likes, actor_3_facebook_likes, actor_2_name, actor_1_facebook_likes, actor_1_name, actor_3_name,-facenumber_in_poster, actor_2_facebook_likes, movie_imdb_link foram extraídas de forma determinística por escolha dos autores deste.
O motivo de remover as variáveis em (1) é devido à tentativa de reduzir o número de missing na base de dados. no passo 3.1 de Tratamento de NA’s.
#db=RECOMMEND.METADATA
db = db %>% select(- director_facebook_likes, -actor_3_facebook_likes, -actor_2_name, -actor_1_facebook_likes, -actor_1_name, -actor_3_name,-facenumber_in_poster, -actor_2_facebook_likes, -color, -director_name, -language, -country, -content_rating, -movie_imdb_link)## O banco de dados agora possui 14 variáveis. Ou seja, foram removidas 14 variáveis no passo anterior.Estrutura dos dados
glimpse(db) #str(db)## Observations: 5,043
## Variables: 14
## $ num_critic_for_reviews <int> 723, 302, 602, 813, NA, 462, 392, 32...
## $ duration <int> 178, 169, 148, 164, NA, 132, 156, 10...
## $ gross <int> 760505847, 309404152, 200074175, 448...
## $ genres <fct> Action|Adventure|Fantasy|Sci-Fi, Act...
## $ movie_title <fct> Avatar , Pirates of the Caribbean: A...
## $ num_voted_users <int> 886204, 471220, 275868, 1144337, 8, ...
## $ cast_total_facebook_likes <int> 4834, 48350, 11700, 106759, 143, 187...
## $ plot_keywords <fct> avatar|future|marine|native|parapleg...
## $ num_user_for_reviews <int> 3054, 1238, 994, 2701, NA, 738, 1902...
## $ budget <dbl> 237000000, 300000000, 245000000, 250...
## $ title_year <int> 2009, 2007, 2015, 2012, NA, 2012, 20...
## $ imdb_score <fct> 7.9, 7.1, 6.8, 8.5, 7.1, 6.6, 6.2, 7...
## $ aspect_ratio <fct> 1.78, 2.35, 2.35, 2.35, , 2.35, 2.35...
## $ movie_facebook_likes <int> 33000, 0, 85000, 164000, 0, 24000, 0...3.2 Transformação de variáveis
Transformando as variáveis imdb_score e aspect_ratio em numéricas
db$imdb_score = as.numeric(as.character(db$imdb_score))
db$aspect_ratio = as.numeric(as.character(db$aspect_ratio))hc1 = hchart(db$imdb_score, color = "#e8bb0b", name = "imdb_score") %>%
hc_title(text = "Histograma dos votos na plataforma IMDB") %>%
hc_exporting(enabled = TRUE, filename = "Fig1-Pimenta"); hc1hc2 = hchart(db$num_voted_users, color = "#786eea", name = "imdb_num_votos") %>%
hc_title(text = "Histograma dos número de votos na plataforma IMDB") %>%
hc_exporting(enabled = TRUE, filename = "Fig2-Pimenta"); hc2Criando uma função para cálculo da moda
Moda <- function(x) {
ux <- unique(x)
ux[which.max(tabulate(match(x, ux)))]
}## Média, moda e mediana da variável imdb_score são, respectivamente, 6.44 6.7 6.6Extraindo valores da variável gênero e transformando em dummies
gg <- as.character(db$genres); gg <- gsub("-", "_", as.character(gg))
#t <- unlist(strsplit(gg[1],split = "\\|"))
tem1 <- data.frame()
for(i in 1:length(gg)){
tem <- tem1
t <- unlist(strsplit(gg[i],split = "\\|"))
temp <- data.frame(t)
tem1 <- rbind(tem,temp)
}
Gen <- unique(tem1); Gen <- gsub("-", "_", as.character(Gen$t))
cat("Existem", length(Gen), "valores de gêneros únicos de filme no banco de dados.")## Existem 26 valores de gêneros únicos de filme no banco de dados.Genname <- Gen
fe <- matrix(data = 0, nrow = length(gg), ncol = length(Genname))
fe <- data.frame(fe); colnames(fe) <- Genname
i=j=0
for(i in 1:length(gg)){
for(j in 1:length(Genname)){
g <- grepl(Genname[j], gg[i])
if(g == TRUE){
fe[i, j] <- 1
}
}
}
NumGen = as_tibble(rbind(apply(fe,2,sum)))
NumGen = gather(NumGen, key = "variables", value = "num_gender")
NumGen = NumGen[order(NumGen$num_gender, decreasing = TRUE), ]
hc3 <- highchart() %>%
hc_add_series(data = NumGen$num_gender,
type = "bar",
name = "# de filmes",
showInLegend = FALSE,
tooltip = list(valueDecimals = 0, valuePrefix = "", valueSuffix = ""), color="blue") %>%
hc_yAxis(title = list(text = "Quantitativo de filmes"),
allowDecimals = TRUE, max = (max(NumGen$num_gender)+103),
labels = list(format = "{value}")) %>%
hc_xAxis(title = list(text = "Gênero de filme"),
categories = NumGen$variables,
tickmarkPlacement = "on",
opposite = FALSE) %>%
hc_title(text = "Quantitativo de filmes por gênero",
style = list(fontWeight = "bold")) %>%
hc_subtitle(text = paste("")) %>%
hc_tooltip(valueDecimals = 2,
pointFormat = "{point.y} filmes")%>%
#pointFormat = "Variável: {point.x} <br> Missing: {point.y}")
hc_credits(enabled = TRUE,
text = "Fonte: IMDB/KAGGLE. Elaboração: Ewerson Pimenta.",
style = list(fontSize = "10px")) %>%
hc_exporting(enabled = TRUE, filename = "F3-filmes-genero-Pimenta")
#hc <- hc %>%
# hc_add_theme(hc_theme_darkunica())
hc3; remove(gg,t,tem, temp, tem1,Gen, g, i, j)Algumas variáveis de gêneros possuem pouquíssimos filmes serão removidas do banco. São elas Film_Noir, Short, News, Reality_TV e Game_Show.
fe$Film_Noir <- NULL
fe$Short <- NULL
fe$News <- NULL
fe$Reality_TV <- NULL
fe$Game_Show <- NULLUnificando a base de dados com as novas variáveis de gêneros únicos descobertos.
db1 = as_tibble(cbind(as.data.frame(db),fe))
cat("Foram inseridas ", dim(fe)[2],"novas variáveis provenientes dos gêneros únicos descobertos no passo anterior.")## Foram inseridas 21 novas variáveis provenientes dos gêneros únicos descobertos no passo anterior.Semelhante à variável de gênero, foi feito o split por palavra chave da variável keyword, esse, por sua vez, continha 6780 palavras únicas. E então, geramos uma nuvem de palavras chave, com o auxílio do pacote wordcloud2 com o seguinte comando, após obter o data frame do nome das palavras chaves e frequência.
library(wordcloud2)
wordcloud2(NumKeyWord)
3.3 Filtrando filmes a partir de 1980
Excluindo filmes lançados antes de 1980
hc00 = hchart(db$title_year, color = "#53a074", name = "imdb1_score") %>%
hc_title(text = "Histograma do ano de publicação dos filmes") %>%
hc_exporting(enabled = TRUE, filename = "Fig1-Pimenta"); hc00Percebemos pelo gráfico anterior que existem poucos filmes publicados antes de 1980 e estes podem não ser representativos. Decidimos, então, por trabalhar apenas com os filmes que foram publicados a aprtir de 1980
db <- db[db$title_year >= 1980,]
hc01 = hchart(db$title_year, color = "#79d8a2", name = "imdb1_score") %>%
hc_title(text = "Histograma do ano de publicação dos filmes") %>%
hc_exporting(enabled = TRUE, filename = "Fig1-Pimenta"); hc013.4 Tratamento de NA, zeros e duplicatas
Porcentagem de NA por variável
db_miss <- db %>% summarise_all(funs(sum(is.na(.))/n()))## Warning: package 'bindrcpp' was built under R version 3.5.1db_miss <- gather(db_miss, key = "variables", value = "percent_missing")
db_miss$percent_missing = 100*db_miss$percent_missing
db_miss = db_miss[order(db_miss$percent_missing, decreasing = TRUE), ]
#db_miss
hc4 <- highchart() %>%
hc_add_series(data = db_miss$percent_missing,
type = "bar",
name = "Porcentagem de missing",
showInLegend = FALSE,
tooltip = list(valueDecimals = 2, valuePrefix = "", valueSuffix = " %")) %>%
hc_yAxis(title = list(text = "Porcentagem de missing"),
allowDecimals = TRUE, max = 20,
labels = list(format = "{value}%")) %>%
hc_xAxis(categories = db_miss$variables,
tickmarkPlacement = "on",
opposite = FALSE) %>%
hc_title(text = "Porcentagem de missinig por variável",
style = list(fontWeight = "bold")) %>%
hc_subtitle(text = paste("")) %>%
hc_tooltip(valueDecimals = 2,
pointFormat = "Missing: {point.y}")%>%
#pointFormat = "Variável: {point.x} <br> Missing: {point.y}")
hc_credits(enabled = TRUE,
text = "Fonte: IMDB/KAGGLE. Elaboração: Ewerson Pimenta.",
style = list(fontSize = "10px")) %>%
hc_exporting(enabled = TRUE, filename = "Fig0-Pimenta")
#hc <- hc %>%
# hc_add_theme(hc_theme_darkunica())
hc4Removendo todas as observações que contêm NA.
db1 <- db1 %>% drop_na(gross, budget, aspect_ratio
,title_year
,num_critic_for_reviews, num_user_for_reviews)
#glimpse(db1)
db1 <- db1 %>% na.omit()
db1 <- db1 %>% drop_na()Removendo dados duplicados
cat("Existem", sum(duplicated(db1)), "filmes duplicados na base de dados.")## Existem 33 filmes duplicados na base de dados.db1 <- db1[!duplicated(db1), ]## O novo banco de dados, sem observações faltantes, possui 3782 observações. Ou seja, no processo de remoção de valores faltantes foram perdidas 1000 observações. Finalmente, removemos todas as observações duplicadas e faltantes do banco.Porcentagem de ZEROS por variável
zeros <- (colSums(db1==0)/nrow(db1)*100); var <- names(zeros)
db_zero <- data.frame(var,zeros); rownames(db_zero) <- NULL
db_zero <- db_zero[order(db_zero$zeros, decreasing = TRUE), ]
hc4_1 <- highchart() %>%
hc_add_series(data = db_zero$zeros,
type = "bar",
name = "Porcentagem de zeros",
showInLegend = FALSE,
tooltip = list(valueDecimals = 2, valuePrefix = "", valueSuffix = " %"), color="pink") %>%
hc_yAxis(title = list(text = "Porcentagem de zero"),
allowDecimals = TRUE, max = 100,
labels = list(format = "{value}%")) %>%
hc_xAxis(categories = db_zero$var,
tickmarkPlacement = "on",
opposite = FALSE) %>%
hc_title(text = "Porcentagem de zeros por variável",
style = list(fontWeight = "bold")) %>%
hc_subtitle(text = paste("")) %>%
hc_tooltip(valueDecimals = 2,
pointFormat = "Zeros: {point.y}")%>%
#pointFormat = "Variável: {point.x} <br> Missing: {point.y}")
hc_credits(enabled = TRUE,
text = "Fonte: IMDB/KAGGLE. Elaboração: Ewerson Pimenta.",
style = list(fontSize = "10px")) %>%
hc_exporting(enabled = TRUE, filename = "Fig00-Pimenta")
#hc <- hc %>%
# hc_add_theme(hc_theme_darkunica())
hc4_1Aqui podemos perceber que os zeros estão concentrados nas variáveis de gênero que são binárias e não faz sentido analizar os zeros nessas variáveis.
paste0("Entretanto, pode-se perceber uma proporção de zeros na variável movie_facebook_likes igual a ",round(db_zero[22,2],2), "% e, também na variável cast_total_facebook_likes de ",round(db_zero[23,2],2),"%. Decidimos por remover a variável movie_facebook_likes.")## [1] "Entretanto, pode-se perceber uma proporção de zeros na variável movie_facebook_likes igual a 46.17% e, também na variável cast_total_facebook_likes de 0.13%. Decidimos por remover a variável movie_facebook_likes."db1$movie_facebook_likes <- NULL3.5 Weighted Rating (WR) - IMDB_score
Um passo importante é penalizar a variável de escore imdb_score pelo número de votos recebidos. Ver mais no estimador de Shrinkage
\(WR = \frac{v}{v+m} \times R + \frac{m}{v+m} \times C\)
Onde,
R = as.numeric(db1$imdb_score)
v = as.numeric(db1$num_voted_users)
m = summary(db1$num_voted_users)[2] # 1st Qu.
C = mean(db1$imdb_score)
db1$WR = (v/(v+m))*R + (m/(v+m))*C- R = Escore médio dos votos para o título do filme dado pelos usuários do IMDB = (imdb_score)
- v = Número de usuários que votaram = (num_voted_users)
- m = Mínimo de votos requerido (atualmente 7.000)
- C = O escore médio de todos os 3766 filmes (atualmente 6,5)
Medidas pontuais e de dispersão de imdb_score e WR
summary(db1$imdb_score)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.600 5.900 6.600 6.468 7.200 9.300summary(db1$WR)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.578 6.166 6.498 6.573 6.983 9.269cat("Além disso, conseguimos reduzir o desvio padrão que orignalmente era de ",round(sd(db1$imdb_score),2),",para a variável imdb_score, e agora passou a ser ",round(sd(db1$WR),2), "para a variável WR.")## Além disso, conseguimos reduzir o desvio padrão que orignalmente era de 1.05 ,para a variável imdb_score, e agora passou a ser 0.72 para a variável WR.Selecionando uma amostra aleatória (a.a.) de tamanho \(n=800\) e representando em um gráfico de dispersão com valores reais vs ajustados.
set.seed(123654)
amostra0 = sample(x = 1:dim(db1)[1], size = 800, replace = FALSE)
dbX = db1[amostra0,] %>%
select(movie_title,num_voted_users,imdb_score, WR)
dss <- map(c("cross"), function(s){
x <- as.numeric(dbX$imdb_score)
y <- as.numeric(dbX$WR)
list(name = s,
data = list_parse(data_frame(x, y)),
marker = list(symbol = s, enabled = TRUE), lineColor = "#56667a")
})
#dss[[1]]$data[amostra1]
hc5 = highchart() %>%
hc_chart(type = "scatter", color = "#56667a") %>%
hc_title(text = "Score IMDB vs WR (calculado pelo estimador de Shrinkage)") %>%
hc_subtitle(text = "800 filmes selecionados via amostra aleatória simples") %>%
hc_xAxis(title = list(text = "x: imdb_score"),
allowDecimals = TRUE, labels = list(format = "{value}★")) %>%
hc_yAxis(title = list(text = "y: WR (calibrado)"),
allowDecimals = TRUE, labels = list(format = "{value}★")) %>%
hc_exporting(enabled = TRUE, filename = "F3-Pimenta") %>%
hc_add_series_list(dss); hc5; remove(dbX, dss)Análise do cálculo de IMDB
- Para filmes com # de votos recebidos MENOR que 18mil (m: votos mínimos requeridos):
- imdb_score > 6,5: DECRESCIMENTO
- imdb_score < 6,5: CRESCIMENTO
- imdb_score > 6,5: DECRESCIMENTO
- Para filmes com # de votos recebidos MAIOR que 18mil (m: votos mínimos requeridos):
- imdb_score > 6,5: DECRESCIMENTO
- imdb_score < 6,5: CRESCIMENTO
- imdb_score > 6,5: DECRESCIMENTO
Ou seja, para os filmes catalogados com m muito inferior a 18mil o novo escore calibrado teve maior diferentça que aqueles superior a 18 mil. Além disso, quando scores são maiores que 6,5 o WR tende a cair, caso contrário o valor pode aumentar. Quanto maior o número de votos recebidos, menor a diferença do valor de IMDB_score e WR.
set.seed(123654)
amostra2 = sample(x = 1:dim(db1)[1], size = 35, replace = FALSE)
db1[amostra2,] %>%
select(movie_title,num_voted_users,imdb_score, WR, budget) %>%
datatable(rownames = amostra2,options = list(searchin = TRUE, scrollX = TRUE, pageLength = 5))3.6 Visualizações Finais
Finalmente, segue uma a.a.da base de dados após limpeza e análise exploratória.
set.seed(123851)
amostra3 = sample(x = 1:dim(db1)[1], size = 35, replace = FALSE)
db1[amostra3,] %>%
datatable(rownames = amostra3,options = list(searchin = TRUE, scrollX = TRUE, pageLength = 5))Para os dados quantitativos a seguinte Matriz de correlação. Foi gerada utilizando o pacote corrplot
#install.packages("corrplot")
library(corrplot)
res1 <- cor.mtest(db_cor, conf.level = .95)
res2 <- cor.mtest(db_cor, conf.level = .99)
corrplot::corrplot(cor(db_cor), method = "color", col = col(200),
type = "upper", order = "hclust", number.cex = .7,
addCoef.col = "black", # Add coefficient of correlation
tl.col = "black", tl.srt = 90, # Text label color and rotation
# Combine with significance
p.mat = res1$p, sig.level = 0.01, insig = "blank",
# hide correlation coefficient on the principal diagonal
diag = FALSE)
4 Modelagem
4.1 Random Forest - Metodologia
Descrição 1. Random Forest foi desenvolvido para agregar árvores de decisão (modelo de classificação);
2. Pode ser usado para modelo de classificação (p/ var. resposta categórica) ou regressão (no caso de haver variável resposta contínua);
3. Evita overfitting;
4. Permite trabalhar com um largo número de características de um conjunto de dados;
5. Auxilia na seleção de variáveis baseada em um algoritmo que calcula a importância por variável (assim, tendo conhecimento de quais variáveis são mais importantes, podemos usar essa informação para outros modelos de classificação);
6. User-friendly: apenas 2 parâmetros livres:
- Trees - ntrees, default 500 (Nº de árvores);
- Variáveis selecionadas via amostragem aleatória candidatas à cada “split” (quebra da árvore) - mtry, default \(\sqrt{p}\) p/ classificação e \(\frac{p}{3}\) p/ regressão (p: nº de features/variáveis);
Passo-a-Passo
É realizado em 3 passos:
- Desenha as amostras via bootstrap do número de árvores ntrees;
- Para cada amostra via bootstrap, cresce o número de árvores “un-puned” para a escolha da melhor quebra da árvore baseado na amostra aleatória do valor predito de mtry a cada nó da árvore;
- Faz classificação de novos valores usando a maioria de votos p/ classificação e usa a média p/ regressão baseada nas amostras de ntrees.
Exemplo
4.2 Preparação de dados
Plot da variável imdb_score
plot(db1$imdb_score, ylim = c(0,10))
hist(db1$imdb_score)
Plot da variável WR (imdb_score penalizada)
plot(db1$WR, ylim = c(0,10))
hist(db1$WR)
dist1 = 2.5
db2 = db1 %>% filter(WR >= (mean(db1$WR)- dist1) & WR <= (mean(db1$WR)+ dist1))
paste('Excluímos ', dim(db1)[1] - dim(db2)[1], ' filmes da base. O critério aqui utilizado foi o de permanecer, apenas, filmes distantes em até ', dist1, ' estrelas da média de estrelas ',round(mean(db1$WR),2),'.')## [1] "Excluímos 16 filmes da base. O critério aqui utilizado foi o de permanecer, apenas, filmes distantes em até 2.5 estrelas da média de estrelas 6.57 ."Transformação da variável contínua WRem categórica WR_Grp
Transformando a variável WR em fator, com as seguintes categorias:
- Categoria ‘Baixo índice de aprovação (low score)’: [0, 6.5) ★
- Categoria ‘Alto índice de aprovação (high score)’: [6.5, 10) ★
movie = db2
Grp <- function(tn){
tn = abs(tn)
if (tn >= 0 & tn < median(db2$WR)){
return(paste0('[0, ',round(median(db2$WR),2),')'))
}else if(tn >= median(db2$WR)){
return(paste0('[',round(median(db2$WR),2),', 10]'))
}
}
# apply the Group function to the WR column
movie$WR_Grp <- sapply(movie$WR,Grp)
# set as factor the new column
movie$WR_Grp <- as.factor(movie$WR_Grp)
#View(head(movie))
table(movie$WR_Grp)##
## [0, 6.5) [6.5, 10]
## 1883 1883# apply the Group function to the WR column
imdb_score_Grp <- sapply(movie$imdb_score,Grp)
# set as factor the new column
imdb_score_Grp <- as.factor(imdb_score_Grp)
#View(head(movie))
table(imdb_score_Grp)## imdb_score_Grp
## [0, 6.5) [6.5, 10]
## 1696 2070paste('O critério de seleção do valor de dist')## [1] "O critério de seleção do valor de dist"Remoção de variáveis Remove as variáveis imdb_score, genres, plot_keywords, movie_imdb_link e WR.
movie$imdb_score <- NULL
movie$genres <- NULL
movie$plot_keywords <- NULL
movie$movie_imdb_link <- NULL
movie$WR <- NULLglimpse(movie)## Observations: 3,766
## Variables: 32
## $ num_critic_for_reviews <int> 723, 302, 602, 813, 462, 392, 324, 6...
## $ duration <int> 178, 169, 148, 164, 132, 156, 100, 1...
## $ gross <int> 760505847, 309404152, 200074175, 448...
## $ movie_title <fct> Avatar , Pirates of the Caribbean: A...
## $ num_voted_users <int> 886204, 471220, 275868, 1144337, 212...
## $ cast_total_facebook_likes <int> 4834, 48350, 11700, 106759, 1873, 46...
## $ num_user_for_reviews <int> 3054, 1238, 994, 2701, 738, 1902, 38...
## $ budget <dbl> 237000000, 300000000, 245000000, 250...
## $ title_year <int> 2009, 2007, 2015, 2012, 2012, 2007, ...
## $ aspect_ratio <dbl> 1.78, 2.35, 2.35, 2.35, 2.35, 2.35, ...
## $ Action <dbl> 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, ...
## $ Adventure <dbl> 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, ...
## $ Fantasy <dbl> 1, 1, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, ...
## $ Sci_Fi <dbl> 1, 0, 0, 0, 1, 0, 0, 1, 0, 1, 1, 0, ...
## $ Thriller <dbl> 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Documentary <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Romance <dbl> 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, ...
## $ Animation <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, ...
## $ Comedy <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, ...
## $ Family <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, ...
## $ Musical <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, ...
## $ Mystery <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, ...
## $ Western <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Drama <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ History <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Sport <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Crime <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Horror <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ War <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Biography <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...
## $ Music <dbl> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, ...
## $ WR_Grp <fct> [6.5, 10], [6.5, 10], [6.5, 10], [6....cat("A base de dados que iremos trabalhar no modelo tem", dim(movie)[1] , "observações e ", dim(movie)[2]-1, "variáveis, sem contar a variável que contém os títulos dos filmes.")## A base de dados que iremos trabalhar no modelo tem 3766 observações e 31 variáveis, sem contar a variável que contém os títulos dos filmes.Partição dos dados
Particionando a base de dados em Treino e Teste, esses dois (Treino e Teste) também terão armazenos os nomes dos filmes selecionados via amostragem probabilística dos dados originais separadamente das bases de Treino e Teste.
Posteriormente, removemos os rótulos dos filmes nas bases Treino e Teste
set.seed(9182345)
ind <- sample(2, nrow(movie), replace = T, prob = c(0.7, 0.3))
train <- movie[ind==1, -4]
test <- movie[ind==2, -4]
trainMovie <- movie[ind==1, 4]
testMovie <- movie[ind==2, 4]Distribuição dos escores nas bases de treino e teste
round(table(train$WR_Grp)/sum(table(train$WR_Grp))*100,2)##
## [0, 6.5) [6.5, 10]
## 49.79 50.21round(table(test$WR_Grp)/sum(table(test$WR_Grp))*100,2)##
## [0, 6.5) [6.5, 10]
## 50.5 49.54.3 Random Forest - Aplicação e Resultados
Inicialmente utilizaremos o pacote randomForest que implmenta o algoritmo de Random Forest de Breiman (baseado na clusterização de Breiman, originalmente codificada em Fortran) que tem por finalidade classificar e/ou criar regressão. Além disso, pode ser usado em um modelo não supervisionado para avaliar proximidades entre pontos.
Estamos usando, a partir daqui, a base de treino.
#library(randomForest)
#library(rpart)
#library(rpart.plot)
#rf <- randomForest(proximity = T,ntree = 38,do.trace = T,WR~.,data=training)
set.seed(9984512)
rf <- randomForest(WR_Grp~.,data=train)
rf##
## Call:
## randomForest(formula = WR_Grp ~ ., data = train)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 5
##
## OOB estimate of error rate: 18.6%
## Confusion matrix:
## [0, 6.5) [6.5, 10] class.error
## [0, 6.5) 1093 235 0.1769578
## [6.5, 10] 261 1078 0.1949216attributes(rf)## $names
## [1] "call" "type" "predicted"
## [4] "err.rate" "confusion" "votes"
## [7] "oob.times" "classes" "importance"
## [10] "importanceSD" "localImportance" "proximity"
## [13] "ntree" "mtry" "forest"
## [16] "y" "test" "inbag"
## [19] "terms"
##
## $class
## [1] "randomForest.formula" "randomForest"Olhando as 6 primeiras observações real X predito
p1 <- predict(rf,train)
head(p1)## 1 2 3 4 5 6
## [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [0, 6.5)
## Levels: [0, 6.5) [6.5, 10]head(train$WR_Grp)## [1] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [0, 6.5)
## Levels: [0, 6.5) [6.5, 10]Matriz de confusão
library(caret)
library(e1071)## Warning: package 'e1071' was built under R version 3.5.1confusionMatrix(p1, train$WR_Grp)## Confusion Matrix and Statistics
##
## Reference
## Prediction [0, 6.5) [6.5, 10]
## [0, 6.5) 1328 0
## [6.5, 10] 0 1339
##
## Accuracy : 1
## 95% CI : (0.9986, 1)
## No Information Rate : 0.5021
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 1
## Mcnemar's Test P-Value : NA
##
## Sensitivity : 1.0000
## Specificity : 1.0000
## Pos Pred Value : 1.0000
## Neg Pred Value : 1.0000
## Prevalence : 0.4979
## Detection Rate : 0.4979
## Detection Prevalence : 0.4979
## Balanced Accuracy : 1.0000
##
## 'Positive' Class : [0, 6.5)
## 
RF_importance = randomForest::importance(rf)[order(randomForest::importance(rf)[,1], decreasing = TRUE), ]
knitr::kable(RF_importance)| x | |
|---|---|
| num_voted_users | 214.200087 |
| duration | 130.658019 |
| budget | 129.628961 |
| num_user_for_reviews | 128.904402 |
| num_critic_for_reviews | 118.778521 |
| gross | 108.488991 |
| title_year | 88.031601 |
| cast_total_facebook_likes | 86.867504 |
| Drama | 63.647354 |
| Comedy | 23.176435 |
| aspect_ratio | 19.831347 |
| Biography | 19.566187 |
| Thriller | 19.560450 |
| Action | 17.939507 |
| Horror | 17.176821 |
| Animation | 16.687173 |
| Romance | 13.200739 |
| Crime | 11.155917 |
| Adventure | 9.461303 |
| Fantasy | 9.243306 |
| Sci_Fi | 8.888087 |
| Mystery | 7.818714 |
| Family | 7.534521 |
| History | 7.081343 |
| Music | 6.440673 |
| War | 5.641061 |
| Documentary | 4.693869 |
| Musical | 4.294417 |
| Sport | 4.028837 |
| Western | 2.058234 |
| Verificamos que as variávei | Game_Show, Sci_Fi, Reality_TV, News e Film_Noir não foram relevantes para o algoritmo do random forest. |
randomForest::varImpPlot(rf)
Taxa de Erro - Random Forest
plot(rf)
legend('topright', colnames(rf$err.rate), col=1:5, fill=1:5)
Observamos que, a partir do número de árvores geradas \(ntrees > 300\) o erro OOB (Out of Bag) não pode ser melhorado.
Ajustando e melhorando estimativas
Além disso, iremos alterar alguns parâmetros da função randomForest como o número de ntrees e mtry. Assim, repetimos o algoritmo do Random Forest, ainda usando a base treino.
# Tune mtry
x = as.data.frame(train1[,-31])
y = (as.factor(train1$WR_Grp))
t <- tuneRF(x = x, y = y,
stepFactor = 0.3,
plot = TRUE,
ntreeTry = 300,
trace = TRUE,
improve = 0.05)## mtry = 5 OOB error = 17.85%
## Searching left ...
## mtry = 17 OOB error = 18.04%
## -0.0105042 0.05
## Searching right ...
## mtry = 1 OOB error = 24.45%
## -0.3697479 0.05
Aparentemente, 5 é um bom candidato ao valor de \(m_{try}\)
#set.seed(093180)
set.seed(998451)
rf1 <- randomForest(WR_Grp~.,data=train1,
ntree = 300,
mtry = 5,
importance = TRUE,
proximity = TRUE)
rf1##
## Call:
## randomForest(formula = WR_Grp ~ ., data = train1, ntree = 300, mtry = 5, importance = TRUE, proximity = TRUE)
## Type of random forest: classification
## Number of trees: 300
## No. of variables tried at each split: 5
##
## OOB estimate of error rate: 18.45%
## Confusion matrix:
## [0, 6.5) [6.5, 10] class.error
## [0, 6.5) 1105 223 0.1679217
## [6.5, 10] 269 1070 0.2008962RF_importance1 = randomForest::importance(rf1)[order(randomForest::importance(rf1)[,1], decreasing = TRUE), ]
plot(rf1)
legend('topright', colnames(rf$err.rate), col=1:5, fill=1:5)
Removendo variáveis com pouca importância
Decidimos por retirar todos os fatores que retornaram importância abaixo de 10 em MeanDecreaseGini. Portanto, ficaremos apenas com as seguintes variáveis:
RF1 = data.frame(variables = rownames(RF_importance1), importance = RF_importance1[,4])
RF1 = RF1[order(RF1$importance, decreasing = TRUE),]
rownames(RF1) <- NULL
RF1 = RF1[1:10,]
hc6_1 <- highchart() %>%
hc_add_series(data = RF1$importance,
type = "bar",
name = "Importância",
showInLegend = FALSE,
tooltip = list(valueDecimals = 2, valuePrefix = "", valueSuffix = "")) %>%
hc_yAxis(title = list(text = "Importância"),
allowDecimals = TRUE, max = 200,
labels = list(format = "{value}")) %>%
hc_xAxis(title = list(text = "Fatores"),
categories = RF1$variables,
tickmarkPlacement = "on",
opposite = FALSE) %>%
hc_title(text = "Importância por fator - Random Forest",
style = list(fontWeight = "bold")) %>%
hc_subtitle(text = paste("")) %>%
hc_tooltip(valueDecimals = 2,
pointFormat = "Importância: {point.y}")%>%
#pointFormat = "Variável: {point.x} <br> Importância: {point.y}")
hc_credits(enabled = TRUE,
text = "Fonte: IMDB/KAGGLE. Elaboração: Ewerson Pimenta.",
style = list(fontSize = "10px")) %>%
hc_exporting(enabled = TRUE, filename = "F6_1-importance-Pimenta")
#hc <- hc %>%
# hc_add_theme(hc_theme_darkunica())
hc6_1Selecionando variáveis importantes
train2 = train %>% select(num_voted_users, gross, duration, num_user_for_reviews, budget, cast_total_facebook_likes, title_year, num_critic_for_reviews, Drama, Horror, aspect_ratio, Action, Comedy, WR_Grp) %>% droplevels()
test2 = test %>% select(num_voted_users, gross, duration, num_user_for_reviews, budget, cast_total_facebook_likes, title_year, num_critic_for_reviews, Drama, Horror, aspect_ratio, Action, Comedy, WR_Grp) %>% droplevels()
movie = rbind(train2, test2)# Tune mtry
x = as.data.frame(train2[,-14])
y = (as.factor(train2$WR_Grp))
t <- tuneRF(x = x, y = y,
stepFactor = .7,
plot = TRUE,
ntreeTry = 150,
trace = TRUE,
improve = 0.05)## mtry = 3 OOB error = 20.25%
## Searching left ...
## mtry = 5 OOB error = 20.32%
## -0.003703704 0.05
## Searching right ...
## mtry = 2 OOB error = 20.85%
## -0.02962963 0.05
Modelo Final Os parâmetros utilizados, finalmente serão \(m_{try} = 3\) e \(n_{tree} = 150\)
set.seed(093180)
rf_final <- randomForest(WR_Grp~.,data = train2,
ntree = 150,
mtry = 3,
importance = TRUE,
proximity = TRUE)
rf_final; #attributes(rf_final)##
## Call:
## randomForest(formula = WR_Grp ~ ., data = train2, ntree = 150, mtry = 3, importance = TRUE, proximity = TRUE)
## Type of random forest: classification
## Number of trees: 150
## No. of variables tried at each split: 3
##
## OOB estimate of error rate: 20.02%
## Confusion matrix:
## [0, 6.5) [6.5, 10] class.error
## [0, 6.5) 1089 239 0.1799699
## [6.5, 10] 295 1044 0.2203137RF_importance1 = randomForest::importance(rf_final)[order(randomForest::importance(rf_final)[,1], decreasing = TRUE), ]
plot(rf_final)
legend('topright', colnames(rf$err.rate), col=1:5, fill=1:5)
Predição e matriz de confusão - train data
library(caret)
p1 <- predict(rf_final,train2)
head(p1)## 1 2 3 4 5 6
## [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [0, 6.5)
## Levels: [0, 6.5) [6.5, 10]head(train2$WR_Grp)## [1] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [0, 6.5)
## Levels: [0, 6.5) [6.5, 10]confusionMatrix(p1, train2$WR_Grp)## Confusion Matrix and Statistics
##
## Reference
## Prediction [0, 6.5) [6.5, 10]
## [0, 6.5) 1328 0
## [6.5, 10] 0 1339
##
## Accuracy : 1
## 95% CI : (0.9986, 1)
## No Information Rate : 0.5021
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 1
## Mcnemar's Test P-Value : NA
##
## Sensitivity : 1.0000
## Specificity : 1.0000
## Pos Pred Value : 1.0000
## Neg Pred Value : 1.0000
## Prevalence : 0.4979
## Detection Rate : 0.4979
## Detection Prevalence : 0.4979
## Balanced Accuracy : 1.0000
##
## 'Positive' Class : [0, 6.5)
## Predição e matriz de confusão - test data
p2 <- predict(rf_final,test2)
head(p2)## 1 2 3 4 5 6
## [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10]
## Levels: [0, 6.5) [6.5, 10]head(test2$WR_Grp)## [1] [6.5, 10] [0, 6.5) [6.5, 10] [6.5, 10] [6.5, 10] [6.5, 10]
## Levels: [0, 6.5) [6.5, 10]confusionMatrix(p2, test2$WR_Grp)## Confusion Matrix and Statistics
##
## Reference
## Prediction [0, 6.5) [6.5, 10]
## [0, 6.5) 469 105
## [6.5, 10] 86 439
##
## Accuracy : 0.8262
## 95% CI : (0.8025, 0.8482)
## No Information Rate : 0.505
## P-Value [Acc > NIR] : <2e-16
##
## Kappa : 0.6523
## Mcnemar's Test P-Value : 0.1928
##
## Sensitivity : 0.8450
## Specificity : 0.8070
## Pos Pred Value : 0.8171
## Neg Pred Value : 0.8362
## Prevalence : 0.5050
## Detection Rate : 0.4268
## Detection Prevalence : 0.5223
## Balanced Accuracy : 0.8260
##
## 'Positive' Class : [0, 6.5)
## Nº de nós nas árvores
hist(treesize(rf_final),
main = "Nº de nós por ávore",
col = "green")
Extração de uma única árvore \(Árvore \space n_{tree}=1\)
datatable(getTree(rf_final, 1, labelVar = TRUE),
options = list(searchin = TRUE, pageLength = 5))\(Árvore \space n_{tree}=149\)
datatable(getTree(rf_final, 149, labelVar = TRUE),
options = list(searchin = TRUE, pageLength = 5))Gráfico de escala multidimensional da matriz de proximidade Classical multidimensional scaling (MDS) of a data matrix. Also known as principal coordinates analysis (Gower, 1966).
Note that because of numerical errors the computed eigenvalues need not all be non-negative, and even theoretically the representation could be in fewer than n - 1 dimensions.
(MDIM = MDSplot(rf_final, train2$WR_Grp, pch=20))## Warning in RColorBrewer::brewer.pal(nlevs, "Set1"): minimal value for n is 3, returning requested palette with 3 different levels
## $points
## Dim 1 Dim 2
## 1 1.191416e-01 -6.186093e-02
## 2 1.142484e-01 -6.157681e-02
## 3 8.108504e-02 -2.106161e-02
## 4 9.144692e-02 -4.227019e-02
## 5 2.663782e-02 2.323193e-02
## 6 1.892676e-01 -1.251081e-01
## 7 5.039433e-02 3.354186e-03
## 8 1.433929e-01 -8.183954e-02
## 9 1.017028e-01 -4.894726e-02
## 10 2.521623e-02 1.940811e-02
## 11 1.476662e-01 -8.933971e-02
## 12 4.361238e-02 1.008992e-02
## 13 2.014302e-01 -1.319519e-01
## 14 1.228034e-03 4.991470e-02
## 15 1.643644e-01 -1.035877e-01
## 16 1.189432e-01 -5.776495e-02
## 17 3.676391e-02 1.861320e-02
## 18 1.396095e-01 -7.457756e-02
## 19 4.480818e-02 1.136645e-02
## 20 1.366275e-01 -7.233714e-02
## 21 -3.917364e-03 5.210435e-02
## 22 6.662793e-02 -1.345253e-02
## 23 5.591800e-02 -9.949281e-04
## 24 4.410363e-02 1.524389e-02
## 25 1.377117e-01 -7.932194e-02
## 26 1.508044e-01 -9.082310e-02
## 27 7.276718e-03 4.009574e-02
## 28 5.614959e-02 -6.371171e-03
## 29 -2.083999e-03 4.503514e-02
## 30 -1.471098e-02 5.268769e-02
## 31 1.328977e-01 -7.402835e-02
## 32 7.719490e-02 -1.798166e-02
## 33 3.526119e-02 1.313665e-02
## 34 1.533302e-01 -9.026150e-02
## 35 -1.351227e-02 6.097042e-02
## 36 2.565872e-01 -1.603535e-01
## 37 3.265948e-02 2.532414e-02
## 38 1.859393e-01 -1.131938e-01
## 39 5.821899e-02 -4.616075e-03
## 40 9.206719e-02 -3.516168e-02
## 41 7.104558e-02 -1.027434e-02
## 42 -1.200248e-02 5.672862e-02
## 43 1.254094e-01 -6.697627e-02
## 44 7.834424e-02 -2.011233e-02
## 45 -6.388011e-03 5.945485e-02
## 46 -8.449605e-03 5.493368e-02
## 47 -1.756532e-02 6.124555e-02
## 48 1.791718e-02 3.411103e-02
## 49 2.002372e-01 -1.276925e-01
## 50 1.111673e-01 -5.170304e-02
## 51 -1.489558e-02 5.022479e-02
## 52 1.417431e-01 -6.204414e-02
## 53 -3.201112e-02 2.816372e-02
## 54 -1.657322e-02 6.361989e-02
## 55 -2.427544e-02 5.034578e-02
## 56 -7.962686e-03 5.620242e-02
## 57 -3.848377e-03 5.221126e-02
## 58 8.963594e-02 -2.375448e-02
## 59 -5.310054e-03 5.146378e-02
## 60 1.261109e-01 -6.881389e-02
## 61 1.032490e-01 -3.179489e-02
## 62 1.603468e-02 3.358742e-02
## 63 9.270250e-02 -2.965635e-02
## 64 1.754382e-01 -1.086251e-01
## 65 -1.197709e-02 5.353308e-02
## 66 -9.392811e-03 5.413174e-02
## 67 4.419810e-02 1.086372e-02
## 68 5.789006e-02 8.461919e-04
## 69 -4.517950e-03 5.060252e-02
## 70 -1.156088e-02 5.557617e-02
## 71 1.434314e-01 -8.434468e-02
## 72 2.225097e-01 -1.419800e-01
## 73 1.212896e-01 -6.346657e-02
## 74 1.544536e-01 -8.195176e-02
## 75 -2.242924e-02 6.176334e-02
## 76 -7.645760e-03 5.671142e-02
## 77 -2.619643e-02 4.504944e-02
## 78 -1.470034e-02 6.173715e-02
## 79 3.346767e-02 2.037501e-02
## 80 -6.331609e-03 5.391746e-02
## 81 7.969259e-02 -1.441402e-02
## 82 1.719404e-01 -1.117673e-01
## 83 1.716094e-01 -1.059410e-01
## 84 1.259966e-01 -6.279854e-02
## 85 3.427746e-02 1.385185e-02
## 86 -3.132612e-02 2.905351e-02
## 87 1.792946e-01 -1.118041e-01
## 88 1.384445e-01 -7.924413e-02
## 89 1.173732e-01 -6.057641e-02
## 90 2.098671e-01 -1.385296e-01
## 91 9.836321e-02 -2.625196e-02
## 92 1.456981e-01 -8.163663e-02
## 93 -2.388590e-03 5.106083e-02
## 94 -9.162665e-02 -4.343447e-02
## 95 -1.436152e-02 5.721592e-02
## 96 6.330944e-03 5.035400e-02
## 97 -3.772237e-04 4.930809e-02
## 98 -1.448849e-02 6.145782e-02
## 99 9.412741e-02 -3.255662e-02
## 100 2.348878e-02 3.380524e-02
## 101 -1.682248e-01 -1.613760e-01
## 102 -5.038055e-02 6.543911e-03
## 103 -3.822630e-02 3.727722e-02
## 104 -3.818462e-02 2.326085e-02
## 105 -1.714694e-02 5.290451e-02
## 106 -1.450527e-02 5.477573e-02
## 107 6.052131e-02 -1.239089e-03
## 108 4.124574e-02 1.022749e-02
## 109 -8.864281e-03 6.074169e-02
## 110 -2.291057e-02 5.392606e-02
## 111 -1.717481e-02 5.472069e-02
## 112 1.606656e-01 -8.010254e-02
## 113 8.480566e-03 4.001259e-02
## 114 1.944484e-01 -1.304739e-01
## 115 1.359756e-01 -8.038088e-02
## 116 -2.794411e-03 4.818199e-02
## 117 -2.855556e-02 4.979266e-02
## 118 2.326649e-01 -1.515322e-01
## 119 4.020151e-02 1.725374e-02
## 120 -4.095594e-03 5.459501e-02
## 121 -2.335039e-02 5.705361e-02
## 122 -2.701214e-02 5.701717e-02
## 123 3.779956e-03 4.756036e-02
## 124 3.240532e-02 2.643070e-02
## 125 1.148842e-01 -4.864390e-02
## 126 2.140490e-01 -1.329138e-01
## 127 -4.750548e-02 1.990487e-02
## 128 1.591957e-02 3.713941e-02
## 129 -5.417961e-02 5.972931e-03
## 130 1.893454e-01 -1.208273e-01
## 131 1.076677e-02 3.572795e-02
## 132 6.524999e-02 -7.449201e-03
## 133 1.789446e-01 -1.098747e-01
## 134 -1.554388e-02 4.951721e-02
## 135 7.869998e-02 -2.302992e-02
## 136 1.820972e-02 3.742061e-02
## 137 -3.014037e-02 4.406741e-02
## 138 5.666270e-02 3.987693e-04
## 139 -3.861332e-03 5.134750e-02
## 140 1.948995e-01 -1.168937e-01
## 141 -8.275575e-03 5.786551e-02
## 142 -5.961817e-02 -2.065383e-03
## 143 -2.789941e-02 3.662880e-02
## 144 2.522919e-01 -1.643998e-01
## 145 -2.912749e-02 4.984053e-02
## 146 1.331328e-01 -6.490533e-02
## 147 -5.766472e-02 9.898038e-05
## 148 2.270401e-01 -1.543725e-01
## 149 1.406000e-01 -7.031191e-02
## 150 1.177765e-03 4.663514e-02
## 151 7.407913e-02 -3.781878e-03
## 152 4.320676e-02 1.425749e-02
## 153 1.616415e-01 -9.246570e-02
## 154 2.478245e-02 2.270204e-02
## 155 1.603516e-02 3.759506e-02
## 156 -1.749424e-02 5.591669e-02
## 157 -1.696614e-01 -1.688548e-01
## 158 1.732936e-01 -8.956009e-02
## 159 5.748199e-02 5.168726e-03
## 160 3.448700e-02 2.219843e-02
## 161 -1.594862e-02 6.528311e-02
## 162 -1.095712e-02 5.672350e-02
## 163 2.205549e-01 -1.436061e-01
## 164 1.781563e-01 -1.125434e-01
## 165 1.708020e-01 -1.017284e-01
## 166 1.483089e-01 -7.485702e-02
## 167 2.055445e-01 -1.369751e-01
## 168 4.830596e-03 6.058401e-02
## 169 -1.311771e-02 6.021134e-02
## 170 -7.349340e-03 5.287209e-02
## 171 -7.250426e-02 1.172512e-02
## 172 -3.796359e-02 3.505913e-02
## 173 -2.464370e-02 6.425423e-02
## 174 1.311193e-03 4.247815e-02
## 175 4.983124e-03 4.579587e-02
## 176 1.126007e-01 -3.454514e-02
## 177 -7.029860e-03 5.532362e-02
## 178 -8.928012e-02 -3.818729e-02
## 179 -4.194381e-03 5.026452e-02
## 180 -4.029303e-02 2.762719e-02
## 181 -1.347101e-01 -1.250248e-01
## 182 1.255523e-01 -4.732190e-02
## 183 -4.281749e-02 4.115043e-02
## 184 9.501477e-03 4.488109e-02
## 185 1.026864e-01 -4.065665e-02
## 186 2.484652e-01 -1.653155e-01
## 187 -4.597018e-02 4.006510e-02
## 188 2.108282e-01 -1.361977e-01
## 189 1.140655e-01 -3.733630e-02
## 190 -2.687945e-02 5.693690e-02
## 191 -2.274249e-01 -2.497700e-01
## 192 9.444933e-02 -2.201842e-02
## 193 1.880605e-01 -1.114722e-01
## 194 3.068324e-01 -2.179311e-01
## 195 3.759089e-03 4.506104e-02
## 196 1.852368e-01 -1.118038e-01
## 197 1.785938e-01 -8.378517e-02
## 198 -1.852568e-02 6.117138e-02
## 199 -4.199954e-02 2.956702e-02
## 200 3.225733e-02 2.734452e-02
## 201 -2.353318e-02 5.584673e-02
## 202 3.087148e-01 -2.045081e-01
## 203 -5.832878e-04 5.211101e-02
## 204 -4.726584e-02 1.372795e-02
## 205 3.757288e-02 1.596733e-02
## 206 -4.929729e-02 2.420522e-02
## 207 -8.062911e-03 6.161852e-02
## 208 -2.534160e-02 4.606999e-02
## 209 -1.794765e-03 5.101818e-02
## 210 -2.583591e-02 6.554340e-02
## 211 3.453109e-01 -2.076417e-01
## 212 2.125641e-01 -1.269798e-01
## 213 3.568134e-02 2.193818e-02
## 214 -8.651938e-02 -1.974113e-03
## 215 -2.137310e-02 6.317039e-02
## 216 -7.500928e-02 -5.852401e-03
## 217 -1.367062e-02 5.969489e-02
## 218 -6.421348e-02 2.514243e-02
## 219 -1.141488e-02 5.532913e-02
## 220 -1.073934e-02 6.828789e-02
## 221 -3.655626e-02 2.516096e-02
## 222 4.255651e-03 1.103040e-01
## 223 6.992746e-02 -1.065372e-02
## 224 2.674859e-01 -1.536344e-01
## 225 -1.051375e-01 -5.830596e-02
## 226 -4.778660e-02 1.823237e-02
## 227 -8.032220e-03 6.715555e-02
## 228 -2.429622e-01 -2.561370e-01
## 229 -1.501056e-02 5.274867e-02
## 230 -3.078835e-03 5.625014e-02
## 231 5.563134e-02 2.213340e-02
## 232 2.196595e-01 -1.459483e-01
## 233 1.637239e-01 -9.439615e-02
## 234 1.843375e-01 -1.126528e-01
## 235 -9.420186e-02 -1.306865e-02
## 236 -5.775029e-03 5.491021e-02
## 237 7.752285e-02 6.091935e-03
## 238 1.752235e-02 3.294471e-02
## 239 1.580156e-01 -9.371369e-02
## 240 -4.233520e-02 3.830800e-02
## 241 -1.369527e-02 7.088555e-02
## 242 2.535107e-01 -1.753309e-01
## 243 -6.262942e-02 6.073904e-03
## 244 -1.094196e-02 5.953711e-02
## 245 9.920691e-02 -3.967465e-02
## 246 -1.190497e-02 6.557064e-02
## 247 -1.875604e-02 4.097765e-02
## 248 2.416454e-02 3.786539e-02
## 249 -1.841400e-02 4.710936e-02
## 250 -8.112540e-03 4.918771e-02
## 251 1.468977e-01 -5.795909e-02
## 252 2.524524e-01 -1.639287e-01
## 253 4.827698e-03 3.966848e-02
## 254 1.279840e-01 -5.593611e-02
## 255 2.432881e-01 -1.363776e-01
## 256 1.945936e-01 -1.270589e-01
## 257 -4.301026e-02 3.750269e-02
## 258 -1.269596e-01 -1.113467e-01
## 259 7.557430e-02 2.074866e-03
## 260 1.981512e-01 -8.709683e-02
## 261 1.924262e-01 -7.766119e-02
## 262 -2.151978e-02 6.231071e-02
## 263 -1.299244e-02 6.016126e-02
## 264 -5.223979e-02 2.224897e-02
## 265 -3.058521e-02 5.919772e-02
## 266 -5.731012e-02 2.508073e-02
## 267 -1.011678e-02 5.123702e-02
## 268 -1.318484e-02 5.964055e-02
## 269 1.438177e-02 5.364933e-02
## 270 -3.557970e-02 4.986698e-02
## 271 -7.476876e-03 7.102872e-02
## 272 3.621092e-02 4.211682e-02
## 273 1.194901e-01 -1.929698e-02
## 274 1.212398e-02 5.561154e-02
## 275 5.465181e-05 5.271388e-02
## 276 2.928456e-01 -1.958428e-01
## 277 -1.045578e-02 5.340208e-02
## 278 1.939551e-01 -1.188892e-01
## 279 1.569321e-02 4.635253e-02
## 280 2.355345e-01 -1.580152e-01
## 281 -3.217952e-02 5.776851e-02
## 282 -1.431473e-02 5.287951e-02
## 283 -1.411386e-02 5.263872e-02
## 284 -2.178236e-01 -2.289365e-01
## 285 -2.771311e-02 5.682231e-02
## 286 5.376737e-02 8.325783e-03
## 287 9.405953e-02 -1.342001e-02
## 288 9.079456e-02 -2.855762e-02
## 289 1.659253e-01 -9.137059e-02
## 290 1.825062e-01 -7.002661e-02
## 291 -4.713425e-02 1.776912e-02
## 292 2.840423e-03 4.544586e-02
## 293 -1.217248e-01 -9.546562e-02
## 294 1.090629e-01 -1.444886e-02
## 295 -2.282453e-02 5.958478e-02
## 296 1.037721e-02 4.016736e-02
## 297 -2.083833e-02 4.721765e-02
## 298 -1.920985e-01 -1.903758e-01
## 299 -4.379734e-02 1.927724e-02
## 300 1.250426e-01 -4.863867e-02
## 301 -2.347693e-01 -2.606742e-01
## 302 4.193614e-02 1.463623e-02
## 303 2.531368e-01 -1.346293e-01
## 304 -4.507905e-02 3.998901e-02
## 305 8.627133e-02 -2.180921e-02
## 306 2.174067e-01 -1.398206e-01
## 307 4.650361e-02 1.186580e-02
## 308 -3.393935e-02 3.356736e-02
## 309 1.875625e-02 4.318359e-02
## 310 -2.732554e-02 5.974651e-02
## 311 -2.470478e-02 3.673446e-02
## 312 -2.116183e-01 -2.355517e-01
## 313 -5.845212e-02 8.846095e-03
## 314 -2.316570e-03 5.413658e-02
## 315 8.136065e-03 5.243245e-02
## 316 2.833375e-01 -1.945721e-01
## 317 -4.433601e-02 3.233901e-02
## 318 -6.868096e-02 7.713853e-04
## 319 1.313133e-01 -4.497093e-02
## 320 9.399628e-02 -2.014672e-02
## 321 -1.275747e-02 6.347306e-02
## 322 4.783090e-02 2.547971e-02
## 323 9.806875e-02 -2.466264e-02
## 324 -8.082196e-02 -2.822082e-02
## 325 -2.599860e-02 3.503609e-02
## 326 -4.975711e-02 1.600593e-02
## 327 -6.407240e-02 1.059901e-02
## 328 1.925973e-02 4.300269e-02
## 329 -1.238815e-01 -5.664149e-02
## 330 -1.474469e-01 -1.126458e-01
## 331 -5.805058e-02 1.795160e-02
## 332 3.378696e-03 5.761325e-02
## 333 -4.918111e-02 1.470147e-02
## 334 -6.509983e-02 1.008147e-02
## 335 -5.715258e-02 3.488553e-02
## 336 -1.794339e-02 5.955280e-02
## 337 -8.757508e-02 -1.534877e-02
## 338 -6.910792e-02 -7.372562e-03
## 339 -1.359645e-02 5.597029e-02
## 340 3.870836e-02 8.529625e-03
## 341 -2.284167e-02 4.868702e-02
## 342 6.832414e-03 6.445569e-02
## 343 -1.641635e-01 -1.495660e-01
## 344 -4.890821e-02 2.849517e-02
## 345 -6.502442e-02 1.844248e-02
## 346 -1.109337e-01 -5.329416e-02
## 347 3.265443e-01 -2.123688e-01
## 348 2.005233e-01 -1.129196e-01
## 349 -5.772534e-02 1.487098e-02
## 350 -7.486693e-02 -1.297469e-02
## 351 2.052718e-01 -9.669510e-02
## 352 -2.198824e-02 5.681241e-02
## 353 -6.625879e-02 3.757335e-03
## 354 -2.614818e-02 4.664469e-02
## 355 -2.281207e-01 -2.317990e-01
## 356 -6.498578e-02 -1.368477e-02
## 357 -1.059893e-02 5.676231e-02
## 358 7.118978e-02 -1.230272e-02
## 359 2.111428e-01 -1.321994e-01
## 360 2.112238e-01 -1.375501e-01
## 361 1.577271e-01 -8.251644e-02
## 362 6.395930e-03 3.571813e-02
## 363 1.490211e-01 -7.821797e-02
## 364 1.880844e-04 5.403276e-02
## 365 -4.959190e-02 1.301331e-02
## 366 -1.758956e-01 -1.503450e-01
## 367 2.220960e-01 -1.433761e-01
## 368 8.358974e-03 4.053739e-02
## 369 2.993474e-01 -1.785889e-01
## 370 -2.835235e-03 5.211341e-02
## 371 -1.100700e-01 -4.002480e-02
## 372 1.617579e-02 4.225643e-02
## 373 -2.458887e-03 6.745826e-02
## 374 -7.693497e-02 2.079851e-03
## 375 -2.188689e-02 4.869010e-02
## 376 -7.441643e-03 6.924216e-02
## 377 9.014417e-03 5.078803e-02
## 378 1.292701e-02 5.804491e-02
## 379 -9.921995e-02 -3.039076e-02
## 380 -3.055363e-02 5.497330e-02
## 381 -2.941503e-02 5.586147e-02
## 382 5.333950e-02 5.027681e-03
## 383 -4.338370e-03 6.751187e-02
## 384 -9.595889e-02 -4.829840e-02
## 385 -8.907430e-02 -9.421857e-03
## 386 1.405588e-02 5.172413e-02
## 387 -8.775737e-04 5.858058e-02
## 388 -2.944385e-02 4.909149e-02
## 389 -2.643113e-01 -2.946973e-01
## 390 3.371056e-02 2.797251e-02
## 391 1.199030e-01 -5.276945e-02
## 392 -1.857949e-03 6.885969e-02
## 393 -2.151931e-02 6.316995e-02
## 394 -1.083102e-01 -7.081231e-02
## 395 -5.924818e-02 1.402582e-02
## 396 3.521830e-03 4.398835e-02
## 397 3.381798e-01 -2.264415e-01
## 398 1.783424e-01 -9.777534e-02
## 399 1.503219e-01 -5.934242e-02
## 400 1.468148e-03 5.366435e-02
## 401 -4.143163e-03 5.435967e-02
## 402 -6.228237e-02 2.086331e-02
## 403 -4.409395e-04 5.031979e-02
## 404 2.731827e-01 -1.503458e-01
## 405 -1.113374e-02 6.754774e-02
## 406 7.117846e-02 1.234949e-02
## 407 4.008380e-02 3.574260e-02
## 408 5.948292e-02 3.612041e-02
## 409 -4.655242e-03 5.545378e-02
## 410 -1.339388e-02 6.709285e-02
## 411 -7.663916e-02 3.507623e-03
## 412 -2.295474e-02 5.798355e-02
## 413 -1.854993e-01 -1.455468e-01
## 414 -2.032720e-02 6.640875e-02
## 415 -2.064856e-01 -2.056076e-01
## 416 6.498549e-03 4.568062e-02
## 417 -1.232684e-02 7.158673e-02
## 418 -5.826975e-02 2.763815e-02
## 419 -1.339435e-01 -5.867017e-02
## 420 -2.038096e-01 -1.801790e-01
## 421 -4.821283e-02 4.978978e-02
## 422 -2.739708e-01 -3.043935e-01
## 423 -1.121164e-02 6.093217e-02
## 424 -3.631554e-02 6.124170e-02
## 425 -1.237739e-01 -4.405838e-02
## 426 5.019152e-02 1.395308e-02
## 427 3.556780e-02 4.289633e-02
## 428 1.383262e-01 -7.214949e-02
## 429 -4.912638e-02 3.144809e-02
## 430 -1.862457e-02 5.705811e-02
## 431 -8.694950e-04 6.492301e-02
## 432 -4.202404e-02 3.941717e-02
## 433 8.126847e-02 1.747341e-02
## 434 2.948552e-02 3.730801e-02
## 435 -1.962752e-02 6.324220e-02
## 436 -5.038701e-02 3.493306e-02
## 437 5.550951e-03 6.178204e-02
## 438 -1.787928e-01 -1.466358e-01
## 439 2.478955e-01 -1.652881e-01
## 440 4.044298e-02 3.021926e-02
## 441 1.878974e-02 2.972943e-02
## 442 1.448869e-01 -6.013508e-02
## 443 2.265778e-01 -1.492772e-01
## 444 -1.890417e-01 -1.886453e-01
## 445 1.012998e-01 -6.452521e-03
## 446 6.363552e-02 -9.105681e-03
## 447 -2.088168e-02 3.756512e-02
## 448 1.912765e-01 -9.853605e-02
## 449 -3.324019e-03 5.669403e-02
## 450 -3.097958e-02 4.988489e-02
## 451 -4.192229e-02 3.045876e-02
## 452 -7.332917e-02 1.328329e-02
## 453 -1.747529e-01 -1.571141e-01
## 454 3.246433e-02 3.909260e-02
## 455 2.547840e-03 6.724754e-02
## 456 2.514551e-01 -1.316197e-01
## 457 -2.547692e-02 5.694724e-02
## 458 -1.621480e-02 6.071335e-02
## 459 -2.383226e-03 6.228928e-02
## 460 -4.343595e-02 7.501204e-02
## 461 -4.192173e-03 5.043337e-02
## 462 -1.302723e-02 6.912889e-02
## 463 -7.355295e-03 6.586553e-02
## 464 -9.719581e-02 -3.517842e-02
## 465 -4.989917e-02 4.355975e-02
## 466 -1.311784e-02 7.611665e-02
## 467 -4.211838e-02 5.906986e-02
## 468 -3.761648e-02 5.324642e-02
## 469 -3.786025e-03 6.810196e-02
## 470 2.091227e-02 4.871075e-02
## 471 2.110979e-01 -8.013388e-02
## 472 2.295801e-01 -1.498694e-01
## 473 -5.723021e-03 6.787414e-02
## 474 -1.344834e-03 6.208478e-02
## 475 -2.371286e-01 -2.409896e-01
## 476 -3.109892e-02 6.483823e-02
## 477 -1.511216e-02 6.082808e-02
## 478 -7.503500e-02 -1.461289e-02
## 479 -3.182490e-02 3.520352e-02
## 480 1.611900e-02 3.222052e-02
## 481 9.647687e-03 5.311291e-02
## 482 3.356509e-01 -1.969625e-01
## 483 -3.459445e-02 3.178989e-02
## 484 2.567803e-01 -1.619860e-01
## 485 7.136553e-02 -1.594873e-03
## 486 -1.971970e-02 6.119531e-02
## 487 2.759772e-01 -1.759686e-01
## 488 -2.104213e-01 -1.905685e-01
## 489 1.803329e-01 -1.027743e-01
## 490 -9.651378e-02 -3.177700e-02
## 491 -2.611027e-02 4.050266e-02
## 492 -2.012823e-02 5.866612e-02
## 493 -6.722848e-04 6.873868e-02
## 494 1.490032e-02 5.866087e-02
## 495 6.310518e-02 2.943170e-03
## 496 3.221841e-01 -1.976611e-01
## 497 -4.797382e-02 4.567744e-02
## 498 -1.577294e-02 6.618439e-02
## 499 8.610481e-02 -1.596755e-02
## 500 -2.615878e-02 5.633458e-02
## 501 -2.278595e-01 -2.225055e-01
## 502 -1.304437e-02 5.516427e-02
## 503 2.506544e-02 3.415713e-02
## 504 1.709201e-01 -7.811418e-02
## 505 -5.594455e-02 1.887228e-02
## 506 -7.287755e-02 -1.403112e-02
## 507 6.351707e-02 -3.959846e-03
## 508 -2.456700e-02 6.208890e-02
## 509 -3.303039e-02 4.222273e-02
## 510 -3.998682e-02 4.680113e-02
## 511 -1.882437e-01 -1.802985e-01
## 512 -1.857283e-01 -1.505577e-01
## 513 -8.670419e-03 6.183774e-02
## 514 2.035485e-03 4.976497e-02
## 515 -1.127293e-02 6.694584e-02
## 516 5.997623e-03 4.089018e-02
## 517 1.667554e-01 -7.426320e-02
## 518 -5.771870e-02 3.604752e-03
## 519 -5.215784e-02 3.232767e-02
## 520 -1.796151e-01 -1.296535e-01
## 521 -4.277041e-02 3.954288e-02
## 522 -1.303417e-01 -7.101557e-02
## 523 -4.422488e-02 4.087233e-02
## 524 -2.589132e-01 -2.930763e-01
## 525 -1.553976e-01 -9.011977e-02
## 526 -3.979863e-02 4.122858e-02
## 527 -1.065572e-01 -5.204022e-02
## 528 -7.393038e-03 7.246679e-02
## 529 -1.247452e-01 -6.628256e-02
## 530 -1.493827e-01 -1.079521e-01
## 531 -1.750109e-02 7.356432e-02
## 532 -3.062527e-02 6.009602e-02
## 533 -1.944971e-02 6.120756e-02
## 534 2.324695e-01 -1.076715e-01
## 535 -5.692660e-02 5.882984e-02
## 536 -7.273963e-02 4.202082e-03
## 537 -8.873127e-03 6.177506e-02
## 538 -1.705835e-01 -1.132659e-01
## 539 -1.360029e-02 7.601679e-02
## 540 -3.316626e-03 5.859496e-02
## 541 -2.777383e-01 -3.207600e-01
## 542 -2.537759e-02 7.074786e-02
## 543 -1.356086e-02 6.571279e-02
## 544 -1.322373e-01 -6.052432e-02
## 545 -1.198041e-02 8.867629e-02
## 546 1.600002e-01 -9.402005e-02
## 547 -3.583690e-02 4.694508e-02
## 548 -3.879001e-02 3.521893e-02
## 549 -1.916232e-01 -1.535156e-01
## 550 -3.275440e-01 -3.836972e-01
## 551 -7.096652e-02 5.668877e-03
## 552 -3.397136e-02 4.686179e-02
## 553 -2.864156e-02 7.843255e-02
## 554 -1.263634e-02 6.579742e-02
## 555 -7.503538e-02 4.856119e-03
## 556 -8.279200e-02 -4.726449e-03
## 557 -6.999103e-02 1.568439e-02
## 558 -2.825803e-02 5.346471e-02
## 559 -1.245548e-01 -5.285975e-02
## 560 2.537447e-01 -1.549902e-01
## 561 1.653101e-01 -8.805383e-02
## 562 6.283491e-03 5.781724e-02
## 563 3.432697e-02 3.686458e-02
## 564 -8.401733e-02 -3.234942e-02
## 565 -1.123781e-01 -6.316478e-02
## 566 -2.547278e-02 2.768328e-02
## 567 2.789991e-01 -1.633658e-01
## 568 -1.880613e-02 5.287991e-02
## 569 -9.504764e-02 -4.904978e-02
## 570 7.382279e-02 -6.797751e-03
## 571 -2.727414e-02 6.192864e-02
## 572 -1.029613e-02 7.053122e-02
## 573 -4.098002e-03 6.267327e-02
## 574 2.782778e-01 -1.788120e-01
## 575 -3.683454e-02 3.194050e-02
## 576 3.974908e-01 -2.537673e-01
## 577 -6.368310e-02 1.317246e-02
## 578 -6.102428e-02 8.970739e-03
## 579 -4.955049e-02 2.469930e-02
## 580 -6.721401e-02 -2.908168e-03
## 581 -5.226154e-02 5.112959e-02
## 582 7.245747e-02 -3.812834e-03
## 583 -4.720405e-04 6.742700e-02
## 584 -1.295661e-02 4.890052e-02
## 585 -2.422751e-02 6.088475e-02
## 586 -4.187464e-02 3.231776e-02
## 587 -4.937376e-02 4.287283e-02
## 588 -1.619221e-01 -1.195769e-01
## 589 -9.856093e-02 -5.310208e-02
## 590 2.242983e-01 -9.139428e-02
## 591 -4.256989e-02 4.212833e-02
## 592 -2.142706e-02 6.546406e-02
## 593 -8.250843e-03 7.339523e-02
## 594 -5.119713e-02 4.499190e-02
## 595 -1.906042e-03 6.245323e-02
## 596 -1.194617e-02 7.798668e-02
## 597 8.152272e-04 4.963608e-02
## 598 2.992747e-01 -1.676762e-01
## 599 -1.175758e-03 5.958847e-02
## 600 -1.012835e-02 7.177354e-02
## 601 -2.253412e-02 6.536908e-02
## 602 2.187232e-01 -1.012399e-01
## 603 -1.270871e-01 -8.136213e-02
## 604 5.168840e-02 3.304986e-02
## 605 4.785444e-02 3.359981e-02
## 606 2.014076e-01 -8.672290e-02
## 607 -1.168151e-02 6.535399e-02
## 608 3.275930e-01 -2.166623e-01
## 609 1.699958e-03 5.476188e-02
## 610 -3.216914e-03 4.912311e-02
## 611 1.355257e-01 -1.736522e-02
## 612 -2.031408e-02 4.035051e-02
## 613 2.870803e-02 4.202504e-02
## 614 3.026170e-01 -1.493820e-01
## 615 -2.397424e-01 -2.195066e-01
## 616 1.183472e-01 -3.543322e-02
## 617 1.197836e-01 -4.276178e-02
## 618 2.089948e-01 -1.349787e-01
## 619 8.664931e-02 -1.672131e-02
## 620 3.095073e-02 5.113219e-02
## 621 1.112203e-01 -9.712918e-03
## 622 2.006581e-01 -1.287918e-01
## 623 1.661589e-01 -6.288671e-02
## 624 7.125984e-03 4.723647e-02
## 625 -5.011836e-02 3.037520e-02
## 626 1.542582e-01 -8.005909e-02
## 627 5.599532e-02 2.804584e-02
## 628 5.723554e-03 2.726308e-02
## 629 8.184986e-02 -1.892885e-02
## 630 1.745983e-01 -6.772938e-02
## 631 1.058683e-01 -4.133243e-02
## 632 -1.212594e-01 -7.525987e-02
## 633 -2.442483e-02 4.351427e-02
## 634 -6.902094e-02 -4.458613e-03
## 635 3.253319e-01 -1.930761e-01
## 636 -5.052513e-03 5.154238e-02
## 637 2.544177e-01 -1.240232e-01
## 638 -8.325123e-02 -1.132717e-02
## 639 1.663055e-01 -8.374859e-02
## 640 -1.341479e-01 -9.755118e-02
## 641 -1.908054e-01 -1.692017e-01
## 642 2.581579e-01 -1.406782e-01
## 643 5.242410e-02 9.256842e-03
## 644 -8.782505e-02 -1.978145e-02
## 645 2.651712e-01 -1.661302e-01
## 646 -7.797401e-02 -1.354168e-03
## 647 -2.782891e-01 -3.137884e-01
## 648 -2.091697e-01 -1.902940e-01
## 649 -2.117812e-02 6.795119e-02
## 650 2.878640e-02 4.798216e-02
## 651 -5.136336e-04 5.710244e-02
## 652 -1.924612e-01 -1.663548e-01
## 653 -1.597100e-02 7.225544e-02
## 654 4.905670e-02 1.889253e-02
## 655 1.524050e-02 3.227010e-02
## 656 8.089159e-03 6.144363e-02
## 657 -4.605767e-02 3.983170e-02
## 658 -8.274723e-02 -1.995614e-02
## 659 -6.908044e-02 7.574838e-03
## 660 -1.102757e-02 5.536327e-02
## 661 -5.592389e-02 4.390638e-02
## 662 -4.849121e-03 6.839295e-02
## 663 2.282682e-02 5.540138e-02
## 664 -9.144099e-02 -2.657381e-02
## 665 3.657081e-02 5.542018e-02
## 666 -4.414855e-02 4.768233e-02
## 667 -3.487506e-02 5.478832e-02
## 668 -3.405676e-02 5.359617e-02
## 669 -6.501313e-03 6.541780e-02
## 670 -1.543597e-02 5.871809e-02
## 671 -3.258544e-01 -3.743695e-01
## 672 -1.459695e-02 6.737106e-02
## 673 1.322124e-01 -3.952096e-02
## 674 -1.268100e-01 -4.081849e-02
## 675 -3.720022e-03 6.283370e-02
## 676 -4.094943e-02 4.608205e-02
## 677 -3.738032e-02 3.991197e-02
## 678 -2.039904e-02 7.222896e-02
## 679 -5.246208e-02 2.792492e-02
## 680 -1.468625e-02 6.835356e-02
## 681 -1.412477e-02 8.272724e-02
## 682 -4.000871e-03 6.523995e-02
## 683 -3.704312e-02 5.331109e-02
## 684 -7.651672e-03 7.148463e-02
## 685 -2.063207e-02 5.114010e-02
## 686 -7.670479e-02 -1.102318e-02
## 687 -2.503362e-01 -2.757023e-01
## 688 -3.344975e-02 5.455261e-02
## 689 -3.639955e-02 5.252024e-02
## 690 -7.138924e-02 -2.542230e-03
## 691 -1.401550e-01 -8.935872e-02
## 692 1.429138e-01 -3.274183e-02
## 693 1.946572e-01 -1.124886e-01
## 694 6.204902e-02 7.653629e-03
## 695 2.548584e-02 5.724750e-02
## 696 2.246799e-01 -1.099199e-01
## 697 3.791672e-03 5.088946e-02
## 698 -1.167276e-02 6.852321e-02
## 699 -4.870308e-02 1.346921e-02
## 700 9.443995e-02 1.311691e-02
## 701 -3.538110e-02 4.170210e-02
## 702 -1.539819e-01 -1.001333e-01
## 703 -2.085339e-02 6.063683e-02
## 704 -6.665910e-03 5.428103e-02
## 705 2.059526e-02 7.440304e-02
## 706 4.768613e-03 5.562972e-02
## 707 -7.712379e-02 -1.860574e-02
## 708 -3.163371e-02 4.911094e-02
## 709 -2.701644e-01 -3.160131e-01
## 710 5.123284e-03 6.628928e-02
## 711 -3.867928e-02 5.813109e-02
## 712 3.699848e-02 -7.519389e-03
## 713 -1.306356e-02 6.503484e-02
## 714 -1.018489e-01 -6.070219e-02
## 715 5.538256e-02 7.132867e-03
## 716 -5.516324e-02 1.026531e-02
## 717 -3.506456e-02 5.491775e-02
## 718 -1.297024e-01 -9.475565e-02
## 719 -8.503573e-04 7.499888e-02
## 720 7.550788e-02 -5.970173e-03
## 721 -1.573863e-01 -1.076798e-01
## 722 2.833500e-01 -1.588385e-01
## 723 -2.073450e-02 5.746346e-02
## 724 3.724265e-02 1.717457e-02
## 725 -1.559700e-02 5.426616e-02
## 726 6.515258e-03 4.393728e-02
## 727 3.231038e-01 -1.941847e-01
## 728 -3.222202e-02 5.212911e-02
## 729 -2.260951e-02 5.552260e-02
## 730 -2.013397e-01 -2.008024e-01
## 731 7.593035e-03 4.270015e-02
## 732 -2.941189e-01 -3.452033e-01
## 733 -1.772762e-02 6.192638e-02
## 734 -6.197727e-02 1.343343e-02
## 735 -7.851045e-02 -1.174847e-03
## 736 -3.685368e-02 5.997616e-02
## 737 5.071212e-02 3.791194e-02
## 738 -7.585685e-02 4.451355e-03
## 739 -1.225456e-01 -7.735801e-02
## 740 1.024202e-02 5.322106e-02
## 741 7.107484e-02 2.388912e-02
## 742 1.249726e-01 -9.517234e-03
## 743 -3.085993e-03 6.610806e-02
## 744 -9.441787e-02 -9.689354e-03
## 745 -7.152529e-02 1.733384e-02
## 746 -3.381598e-01 -4.163188e-01
## 747 -3.742974e-02 2.366157e-02
## 748 1.037173e-01 -2.729399e-02
## 749 -1.753666e-01 -1.380643e-01
## 750 -9.649384e-03 7.005899e-02
## 751 1.851993e-03 7.031835e-02
## 752 1.040021e-02 4.985463e-02
## 753 -1.358458e-01 -5.486691e-02
## 754 -1.947559e-01 -1.546160e-01
## 755 -7.103109e-03 7.663696e-02
## 756 -1.158760e-01 -5.465018e-02
## 757 6.695969e-02 1.674592e-02
## 758 -6.552904e-02 1.965171e-02
## 759 8.667213e-02 2.604772e-02
## 760 -3.266133e-02 3.326067e-02
## 761 -5.728846e-04 7.757389e-02
## 762 -1.220718e-02 5.830169e-02
## 763 6.385620e-03 7.230568e-02
## 764 -7.761067e-02 -2.821325e-02
## 765 1.588225e-02 9.112067e-02
## 766 -4.488941e-02 4.675503e-02
## 767 -7.604398e-03 4.505824e-02
## 768 -9.586648e-02 -3.638259e-02
## 769 -2.299793e-02 5.679228e-02
## 770 -1.191978e-02 5.522169e-02
## 771 2.120782e-03 9.575933e-02
## 772 -1.634133e-01 -1.228532e-01
## 773 3.531282e-02 5.588003e-02
## 774 -1.331351e-02 5.913915e-02
## 775 -9.035010e-02 -3.237063e-02
## 776 6.147522e-02 1.098237e-02
## 777 -2.721934e-01 -2.708623e-01
## 778 -1.828540e-01 -1.443701e-01
## 779 -7.459240e-02 -2.542263e-02
## 780 1.376835e-02 3.540743e-02
## 781 -2.592453e-01 -2.570413e-01
## 782 2.073217e-01 -1.186243e-01
## 783 -2.728948e-02 5.194049e-02
## 784 -2.142296e-01 -2.222164e-01
## 785 2.329533e-01 -1.454152e-01
## 786 -5.172671e-02 2.165080e-02
## 787 1.374259e-01 -5.860567e-02
## 788 1.656776e-01 -8.985355e-02
## 789 1.352718e-01 -4.941578e-02
## 790 5.225241e-02 5.464905e-02
## 791 1.652357e-02 5.945236e-02
## 792 -1.689116e-02 6.301816e-02
## 793 2.597657e-01 -1.578679e-01
## 794 1.532719e-02 5.204106e-02
## 795 -2.194533e-02 5.815430e-02
## 796 2.524155e-02 6.284369e-02
## 797 1.182900e-01 -2.039008e-04
## 798 -3.431044e-02 4.189735e-02
## 799 1.058571e-02 5.804474e-02
## 800 8.845634e-02 2.805763e-02
## 801 2.658551e-01 -1.646435e-01
## 802 7.711096e-03 5.272185e-02
## 803 3.307072e-02 5.389736e-02
## 804 -3.724693e-02 2.316261e-02
## 805 4.558461e-02 2.081445e-02
## 806 -4.331448e-02 3.953069e-02
## 807 -3.709559e-02 6.933490e-02
## 808 1.003553e-02 5.101369e-02
## 809 -5.658377e-02 3.865939e-02
## 810 -1.102286e-01 -3.918822e-02
## 811 -1.052315e-01 -3.506377e-02
## 812 4.115092e-02 3.923433e-02
## 813 8.758648e-02 2.602323e-03
## 814 1.491996e-01 -5.723103e-02
## 815 2.947611e-01 -1.877711e-01
## 816 -4.158480e-02 4.383590e-02
## 817 -2.392471e-01 -2.699087e-01
## 818 3.521232e-02 2.338689e-02
## 819 -6.151175e-02 2.070008e-03
## 820 -4.142899e-02 6.272887e-02
## 821 -3.874241e-02 5.342947e-02
## 822 -2.203084e-01 -2.056800e-01
## 823 -9.722567e-03 8.276426e-02
## 824 4.776227e-03 1.036841e-01
## 825 -4.598277e-02 6.932429e-02
## 826 3.505452e-03 9.601369e-02
## 827 -1.952590e-02 5.896858e-02
## 828 -3.314271e-02 6.853045e-02
## 829 -3.222059e-02 6.629073e-02
## 830 7.683691e-02 -6.920049e-03
## 831 -5.257799e-02 3.006359e-02
## 832 7.622139e-03 2.195666e-02
## 833 1.361135e-02 1.025604e-01
## 834 -5.262835e-02 6.715481e-02
## 835 -2.827974e-01 -3.108700e-01
## 836 8.805424e-02 3.814801e-02
## 837 -7.235525e-02 8.670290e-03
## 838 -4.743147e-02 4.804073e-02
## 839 1.548625e-02 3.875953e-02
## 840 -1.374425e-01 -1.021777e-01
## 841 -2.031042e-02 6.281629e-02
## 842 -1.277335e-01 -5.660706e-02
## 843 -4.938848e-02 5.312234e-02
## 844 -2.320164e-01 -2.511325e-01
## 845 -1.367830e-01 -7.863309e-02
## 846 -1.902049e-01 -1.604746e-01
## 847 7.087994e-03 7.479351e-02
## 848 -7.815142e-02 2.353817e-02
## 849 2.905330e-02 9.160509e-02
## 850 -1.741053e-02 6.322991e-02
## 851 -6.513498e-03 7.056380e-02
## 852 -2.916866e-02 6.688109e-02
## 853 1.049327e-02 6.634820e-02
## 854 -1.283188e-01 -7.998532e-02
## 855 -4.856432e-02 4.830619e-02
## 856 8.442216e-02 1.648236e-02
## 857 -1.049465e-01 -5.258794e-02
## 858 -2.297840e-01 -2.392015e-01
## 859 -1.259987e-01 -9.276169e-02
## 860 2.548257e-01 -1.654738e-01
## 861 -4.789846e-02 4.203447e-02
## 862 -3.367639e-02 5.554861e-02
## 863 -1.341037e-01 -7.014308e-02
## 864 -1.491742e-01 -8.635388e-02
## 865 -2.796855e-01 -3.247739e-01
## 866 -1.093997e-01 -6.756825e-02
## 867 -3.417684e-02 6.546217e-02
## 868 -3.532524e-02 5.303459e-02
## 869 -2.240878e-01 -2.166253e-01
## 870 -4.468105e-02 3.196122e-02
## 871 -1.887154e-02 6.749413e-02
## 872 -1.251750e-01 -7.772481e-02
## 873 2.513320e-01 -1.164616e-01
## 874 -6.471121e-03 7.130973e-02
## 875 -4.516499e-02 1.451306e-02
## 876 -2.222890e-02 6.157486e-02
## 877 3.546170e-02 3.888987e-02
## 878 1.873712e-02 7.026244e-02
## 879 1.391222e-01 -7.099320e-02
## 880 -3.039815e-02 6.042761e-02
## 881 -4.439095e-02 2.832194e-02
## 882 -1.091744e-02 6.431859e-02
## 883 -9.671449e-02 -3.050935e-02
## 884 -2.251013e-02 3.424576e-02
## 885 -1.856935e-01 -1.638045e-01
## 886 -6.193076e-02 2.054806e-02
## 887 -1.277818e-02 8.082290e-02
## 888 6.003070e-03 8.447041e-02
## 889 -2.662283e-02 6.872677e-02
## 890 -1.229314e-01 -7.069590e-02
## 891 -1.819198e-01 -1.812328e-01
## 892 -4.947667e-02 5.747136e-02
## 893 -7.694808e-02 6.717714e-03
## 894 6.610377e-02 8.802920e-03
## 895 -1.815350e-02 7.248298e-02
## 896 -5.234964e-02 3.541776e-02
## 897 4.967526e-02 8.862327e-02
## 898 -1.497741e-02 7.598350e-02
## 899 -5.343959e-02 1.452499e-02
## 900 -2.552475e-02 5.760593e-02
## 901 1.085310e-01 1.346078e-02
## 902 8.599814e-02 1.300573e-02
## 903 -3.448124e-02 4.951497e-02
## 904 -1.857835e-01 -1.602557e-01
## 905 2.173577e-01 -1.097419e-01
## 906 -2.151370e-02 7.007776e-02
## 907 1.446233e-01 -7.759116e-02
## 908 -1.356107e-01 -8.369300e-02
## 909 -4.053561e-02 5.001684e-02
## 910 -3.535074e-02 3.485134e-02
## 911 -2.408563e-02 5.215706e-02
## 912 -2.227962e-02 6.241564e-02
## 913 -3.877777e-02 5.179726e-02
## 914 -3.620246e-02 5.010552e-02
## 915 -1.546196e-02 7.004723e-02
## 916 2.197757e-01 -9.435701e-02
## 917 -2.085896e-01 -2.141194e-01
## 918 -2.739147e-01 -3.165294e-01
## 919 -3.591382e-01 -4.409193e-01
## 920 -2.687473e-02 5.761396e-02
## 921 4.847939e-02 1.144430e-02
## 922 -4.343885e-02 3.692625e-02
## 923 -1.553461e-02 4.331773e-02
## 924 -8.673614e-02 -2.441163e-02
## 925 1.590634e-01 -8.403889e-02
## 926 2.600370e-01 -1.541100e-01
## 927 1.956773e-02 5.502198e-02
## 928 -2.607423e-02 4.492755e-02
## 929 2.527239e-03 4.746567e-02
## 930 -2.297839e-01 -2.378770e-01
## 931 1.458334e-01 -4.244930e-03
## 932 4.350959e-01 -2.662268e-01
## 933 -1.364163e-01 -8.615125e-02
## 934 -1.174459e-01 -8.316907e-02
## 935 -4.817793e-02 2.828208e-02
## 936 1.100938e-01 -1.779870e-02
## 937 -2.535461e-02 6.015455e-02
## 938 -1.250174e-02 2.979055e-02
## 939 -6.238164e-03 8.827242e-02
## 940 -6.484149e-02 1.199817e-02
## 941 -3.304892e-02 6.109983e-02
## 942 1.719494e-01 -9.324931e-02
## 943 -8.242696e-02 -2.855871e-02
## 944 2.611076e-02 8.086611e-02
## 945 -2.432009e-02 5.080168e-02
## 946 -1.292153e-01 -1.076249e-01
## 947 -2.850135e-01 -3.054444e-01
## 948 2.156908e-01 -7.438451e-02
## 949 1.549817e-01 -6.385349e-02
## 950 1.704985e-02 4.304129e-02
## 951 -3.767143e-02 4.808230e-02
## 952 1.775280e-03 7.087226e-02
## 953 1.040398e-01 -3.293556e-02
## 954 -3.075435e-02 5.531631e-02
## 955 -2.897828e-01 -3.313316e-01
## 956 -1.004222e-01 -3.173483e-02
## 957 4.023059e-01 -2.375857e-01
## 958 -2.333779e-02 6.024326e-02
## 959 -3.839612e-02 6.118120e-02
## 960 -3.009363e-02 4.926863e-02
## 961 -2.898483e-02 6.585916e-02
## 962 9.979914e-03 1.072569e-01
## 963 -1.842813e-01 -1.772843e-01
## 964 -1.085048e-02 9.050040e-02
## 965 3.233341e-02 4.633979e-02
## 966 -5.002954e-05 8.061813e-02
## 967 -6.436037e-02 1.867376e-02
## 968 -1.760767e-02 6.008719e-02
## 969 -2.265460e-01 -2.014816e-01
## 970 -1.147538e-02 6.835715e-02
## 971 -1.608394e-01 -1.259926e-01
## 972 -9.729412e-03 8.467396e-02
## 973 -2.900059e-02 6.387464e-02
## 974 -2.106527e-01 -1.979354e-01
## 975 -2.816273e-01 -3.356352e-01
## 976 -1.303570e-02 6.898134e-02
## 977 -1.813535e-02 6.831120e-02
## 978 -1.633610e-02 7.125392e-02
## 979 -1.862877e-01 -1.588670e-01
## 980 -3.211394e-02 5.968494e-02
## 981 1.625782e-01 -3.150443e-02
## 982 -2.554634e-02 6.419050e-02
## 983 -2.851908e-02 5.696256e-02
## 984 -2.493248e-01 -2.613123e-01
## 985 6.706510e-03 7.623015e-02
## 986 -1.136848e-01 -4.889631e-02
## 987 5.199980e-03 5.184192e-02
## 988 -1.545490e-02 6.579352e-02
## 989 -1.013044e-02 8.404428e-02
## 990 -8.921543e-03 8.716670e-02
## 991 2.093700e-01 -1.122294e-01
## 992 -2.737907e-01 -3.171692e-01
## 993 2.122284e-03 1.104905e-01
## 994 -5.191389e-02 2.753578e-02
## 995 -1.784972e-02 6.025583e-02
## 996 -2.230220e-02 6.119115e-02
## 997 -1.557392e-01 -1.297046e-01
## 998 -8.213021e-03 8.051633e-02
## 999 1.561613e-01 -2.953844e-02
## 1000 -4.669928e-02 3.910775e-02
## 1001 -8.352907e-02 2.034445e-02
## 1002 -3.575454e-02 5.554454e-02
## 1003 6.526820e-03 6.056585e-02
## 1004 -1.371849e-01 -7.453486e-02
## 1005 1.135276e-01 1.323269e-02
## 1006 -1.728786e-03 1.205880e-01
## 1007 -3.904226e-02 4.455216e-02
## 1008 -1.295570e-02 6.772342e-02
## 1009 2.198493e-01 -1.239551e-01
## 1010 2.878747e-01 -1.063412e-01
## 1011 3.596811e-03 1.049710e-01
## 1012 -2.804335e-01 -3.226019e-01
## 1013 -2.306481e-02 6.295833e-02
## 1014 2.508543e-01 -1.471166e-01
## 1015 2.127951e-02 8.166779e-02
## 1016 -1.729637e-01 -1.505284e-01
## 1017 2.699952e-01 -1.667393e-01
## 1018 -1.436147e-02 8.030131e-02
## 1019 5.569063e-02 1.795427e-02
## 1020 -2.010791e-02 6.154715e-02
## 1021 -2.279143e-01 -2.460651e-01
## 1022 -1.522410e-03 6.974192e-02
## 1023 -5.076192e-02 2.147993e-02
## 1024 -9.881355e-02 -5.670749e-02
## 1025 -5.942858e-02 2.839623e-02
## 1026 -1.037721e-01 -7.027047e-02
## 1027 -1.053372e-02 8.825259e-02
## 1028 -1.489396e-02 6.912786e-02
## 1029 -1.644367e-01 -1.332963e-01
## 1030 9.269392e-03 1.141126e-01
## 1031 -2.206940e-02 6.291047e-02
## 1032 2.156600e-02 1.031245e-01
## 1033 2.712862e-01 -1.533721e-01
## 1034 -2.979536e-02 6.623627e-02
## 1035 7.754701e-02 3.978316e-02
## 1036 -1.094418e-02 5.339560e-02
## 1037 -4.061183e-02 3.935748e-02
## 1038 -1.042574e-01 -4.454396e-02
## 1039 2.304924e-02 2.931326e-02
## 1040 2.690476e-02 3.500278e-02
## 1041 1.740431e-01 -7.734342e-02
## 1042 1.923170e-01 -1.089987e-01
## 1043 1.837335e-01 -7.514565e-02
## 1044 2.348003e-02 4.039642e-02
## 1045 1.187700e-02 3.339226e-02
## 1046 6.136488e-02 -5.612061e-03
## 1047 -7.203400e-02 -4.579515e-03
## 1048 -2.169708e-02 1.972683e-02
## 1049 -9.710088e-02 -6.024662e-02
## 1050 4.323494e-01 -2.670208e-01
## 1051 2.885362e-01 -1.823097e-01
## 1052 -1.888680e-01 -1.733587e-01
## 1053 4.183934e-01 -2.182076e-01
## 1054 4.613785e-01 -2.936909e-01
## 1055 -2.268120e-02 3.892912e-02
## 1056 4.721345e-01 -2.985095e-01
## 1057 -4.252527e-02 2.838007e-02
## 1058 -7.735377e-02 3.852388e-03
## 1059 -4.667278e-03 1.000174e-01
## 1060 1.689230e-01 7.554182e-03
## 1061 2.850379e-01 -1.534270e-01
## 1062 3.393624e-01 -2.067798e-01
## 1063 -3.081689e-01 -3.699025e-01
## 1064 -1.845047e-02 6.833968e-02
## 1065 2.825962e-01 -1.306539e-01
## 1066 6.463731e-03 5.605036e-02
## 1067 -1.111624e-01 -6.028377e-02
## 1068 -2.134866e-01 -2.256809e-01
## 1069 -6.008711e-02 6.132437e-03
## 1070 1.688310e-01 -8.290883e-02
## 1071 -1.319213e-01 -1.014011e-01
## 1072 2.851244e-01 -1.024253e-01
## 1073 -1.258056e-02 6.720833e-02
## 1074 -4.525480e-02 2.171532e-02
## 1075 -9.297750e-02 -5.386264e-02
## 1076 2.699489e-02 4.909956e-02
## 1077 1.979977e-01 -2.455552e-02
## 1078 3.777985e-01 -1.778969e-01
## 1079 -3.451299e-02 4.427437e-02
## 1080 -3.784427e-02 3.575328e-02
## 1081 2.559048e-01 -1.281116e-01
## 1082 1.042818e-01 1.921369e-02
## 1083 5.014637e-02 5.960232e-02
## 1084 -1.598135e-02 6.275148e-02
## 1085 1.786429e-01 -7.521012e-02
## 1086 4.310065e-03 4.999529e-02
## 1087 -2.131156e-02 4.178244e-02
## 1088 -3.121305e-02 4.132374e-02
## 1089 -6.172995e-02 -9.931120e-04
## 1090 9.398948e-02 6.634010e-03
## 1091 -9.422076e-02 -2.798755e-02
## 1092 -1.993565e-02 6.642159e-02
## 1093 -3.414501e-01 -4.112845e-01
## 1094 -3.382144e-03 6.122938e-02
## 1095 -1.396382e-01 -1.215793e-01
## 1096 -1.861166e-01 -1.912017e-01
## 1097 -3.062941e-02 5.265445e-02
## 1098 -8.996251e-03 7.872935e-02
## 1099 -8.027244e-02 -2.574263e-03
## 1100 -1.745722e-01 -1.355993e-01
## 1101 -3.174542e-02 4.899431e-02
## 1102 -1.233647e-02 7.800655e-02
## 1103 -6.515043e-02 2.655096e-02
## 1104 -8.714290e-02 -5.644739e-03
## 1105 9.450946e-02 4.312202e-02
## 1106 -7.098324e-02 4.033380e-02
## 1107 2.922386e-02 6.195956e-02
## 1108 3.047200e-03 7.758106e-02
## 1109 -3.376416e-03 7.807759e-02
## 1110 -2.480781e-01 -2.678423e-01
## 1111 -4.886412e-03 7.310577e-02
## 1112 -7.120361e-02 -7.362874e-03
## 1113 4.790035e-03 9.109094e-02
## 1114 -1.664619e-01 -1.105207e-01
## 1115 -6.606066e-03 7.974344e-02
## 1116 -4.602710e-02 4.519474e-02
## 1117 -3.414143e-02 4.599708e-02
## 1118 -1.475002e-02 5.721767e-02
## 1119 -3.982433e-02 5.995012e-02
## 1120 -1.951795e-01 -1.990773e-01
## 1121 -2.594545e-01 -2.995126e-01
## 1122 -1.851097e-02 7.518313e-02
## 1123 1.486173e-02 4.885072e-02
## 1124 -1.018474e-01 -4.557018e-02
## 1125 1.195434e-01 5.611387e-02
## 1126 2.731467e-01 -8.576091e-02
## 1127 -4.098205e-03 1.072815e-01
## 1128 -3.900698e-02 4.849246e-02
## 1129 -1.050829e-01 -5.199812e-02
## 1130 -1.436597e-02 6.046386e-02
## 1131 -1.251518e-02 7.300216e-02
## 1132 -1.495044e-02 5.332652e-02
## 1133 3.172597e-01 -1.583537e-01
## 1134 3.012128e-03 6.927388e-02
## 1135 -1.347994e-01 -1.173726e-01
## 1136 -3.500977e-01 -4.340907e-01
## 1137 -1.270530e-01 -1.052506e-01
## 1138 -8.575557e-03 6.534576e-02
## 1139 -2.158513e-02 6.359248e-02
## 1140 1.252726e-01 -4.679312e-02
## 1141 3.854028e-02 4.626395e-02
## 1142 1.899257e-01 -7.911962e-02
## 1143 -7.079141e-02 2.746211e-03
## 1144 -1.495650e-02 6.679734e-02
## 1145 2.312837e-02 3.279414e-02
## 1146 5.223359e-02 -3.998120e-02
## 1147 3.675234e-03 8.721696e-02
## 1148 1.164134e-01 -3.818362e-02
## 1149 2.717746e-01 -1.232529e-01
## 1150 3.624518e-02 2.058695e-02
## 1151 -4.507504e-02 4.649019e-02
## 1152 -8.844184e-03 6.653147e-02
## 1153 -9.785797e-02 -4.752937e-02
## 1154 1.712126e-01 -7.636376e-02
## 1155 -1.522793e-02 7.198272e-02
## 1156 -8.669371e-02 -4.253653e-02
## 1157 -3.520018e-02 5.080513e-02
## 1158 -1.073681e-01 -1.689831e-02
## 1159 -1.343501e-01 -1.148063e-01
## 1160 3.819191e-01 -1.845006e-01
## 1161 2.979159e-01 -1.811001e-01
## 1162 -4.243194e-02 2.335698e-02
## 1163 -5.380991e-02 3.927951e-02
## 1164 -2.853845e-01 -3.337145e-01
## 1165 -2.894959e-01 -3.333837e-01
## 1166 -1.232291e-01 -8.798127e-02
## 1167 -7.419391e-02 1.640581e-02
## 1168 -7.911442e-02 -1.163767e-02
## 1169 -6.747618e-04 9.271319e-02
## 1170 2.894309e-01 -1.789710e-01
## 1171 -2.422647e-02 5.681717e-02
## 1172 -7.393744e-05 9.136925e-02
## 1173 -8.841851e-03 8.041513e-02
## 1174 -3.583352e-02 3.843696e-02
## 1175 -9.761385e-03 7.051981e-02
## 1176 4.369003e-02 3.041432e-02
## 1177 -1.459116e-01 -1.335250e-01
## 1178 6.894508e-02 2.454258e-02
## 1179 -1.238340e-01 -5.843750e-02
## 1180 3.557927e-01 -2.004740e-01
## 1181 -6.977535e-04 1.000452e-01
## 1182 -1.780969e-02 6.037880e-02
## 1183 -6.283802e-02 2.140755e-02
## 1184 -2.816619e-02 5.390712e-02
## 1185 -3.772550e-02 4.144034e-02
## 1186 -1.552072e-02 5.908861e-02
## 1187 -1.395625e-01 -1.132427e-01
## 1188 -1.770925e-02 6.867670e-02
## 1189 -1.276286e-01 -7.148566e-02
## 1190 -1.925371e-02 6.419954e-02
## 1191 -9.348425e-03 7.493091e-02
## 1192 -7.538288e-02 -1.181142e-02
## 1193 1.787124e-01 -6.074372e-02
## 1194 -4.104128e-02 1.927326e-02
## 1195 -3.071678e-04 6.754736e-02
## 1196 8.408381e-02 5.723562e-02
## 1197 1.265271e-03 7.215541e-02
## 1198 8.219801e-03 9.799785e-02
## 1199 -3.346619e-03 5.109859e-02
## 1200 -4.531359e-02 1.898284e-02
## 1201 -4.771103e-02 6.908982e-02
## 1202 1.090193e-01 8.157981e-03
## 1203 -2.470823e-02 6.590555e-02
## 1204 -3.152410e-01 -3.650182e-01
## 1205 -9.177169e-03 8.990297e-02
## 1206 1.395927e-02 7.028627e-02
## 1207 -2.806586e-01 -3.003882e-01
## 1208 -5.846657e-02 2.054744e-02
## 1209 -1.473200e-02 8.170910e-02
## 1210 -2.394637e-01 -2.666954e-01
## 1211 -1.163348e-01 -6.538597e-02
## 1212 -2.311346e-01 -2.490598e-01
## 1213 -6.598367e-02 3.380196e-03
## 1214 1.712934e-01 1.312852e-02
## 1215 2.319517e-01 -7.678103e-02
## 1216 2.801194e-01 -1.887112e-01
## 1217 -1.953198e-02 9.283101e-02
## 1218 3.822702e-03 8.423755e-02
## 1219 4.715565e-01 -2.613937e-01
## 1220 2.159505e-01 -8.907447e-02
## 1221 -3.754650e-03 1.014966e-01
## 1222 -2.018130e-01 -1.899077e-01
## 1223 -2.396155e-02 6.717587e-02
## 1224 -2.619773e-01 -2.931083e-01
## 1225 -7.644630e-02 -2.564339e-02
## 1226 3.174623e-01 -1.337912e-01
## 1227 4.249230e-01 -2.482647e-01
## 1228 -6.598950e-02 1.259154e-04
## 1229 1.552107e-01 -4.743986e-02
## 1230 1.871355e-01 -2.618049e-02
## 1231 -1.518435e-01 -1.251840e-01
## 1232 5.722423e-03 4.144214e-02
## 1233 2.519915e-02 6.188494e-02
## 1234 -3.280908e-02 3.315977e-02
## 1235 -4.359277e-02 7.199801e-02
## 1236 -2.103099e-02 6.391153e-02
## 1237 5.309177e-02 2.744901e-02
## 1238 2.480810e-01 -1.119537e-01
## 1239 2.605463e-01 -1.312538e-01
## 1240 -1.753496e-02 7.305405e-02
## 1241 4.474723e-01 -2.459669e-01
## 1242 -2.974094e-01 -3.319485e-01
## 1243 4.594458e-01 -2.460079e-01
## 1244 -1.966586e-01 -1.837357e-01
## 1245 -9.314564e-03 6.498739e-02
## 1246 -1.536578e-01 -1.309862e-01
## 1247 4.234194e-01 -2.408254e-01
## 1248 -3.034367e-01 -3.449286e-01
## 1249 -3.797531e-02 6.075313e-02
## 1250 -1.322472e-01 -8.378165e-02
## 1251 -3.159047e-03 9.067601e-02
## 1252 -9.329882e-04 7.380020e-02
## 1253 -4.330935e-02 3.375371e-02
## 1254 -2.350289e-01 -2.269259e-01
## 1255 -3.467505e-02 5.631845e-02
## 1256 1.116165e-01 2.873891e-02
## 1257 -8.154785e-02 -2.401711e-02
## 1258 -2.322885e-01 -2.183199e-01
## 1259 -1.049753e-02 7.613853e-02
## 1260 4.827049e-02 9.266359e-02
## 1261 -7.792496e-02 -1.608921e-02
## 1262 -5.643808e-03 8.414899e-02
## 1263 -1.298027e-01 -1.076246e-01
## 1264 -2.974107e-01 -3.413219e-01
## 1265 -2.695757e-01 -3.051128e-01
## 1266 -1.941054e-02 7.364267e-02
## 1267 6.538857e-03 8.893804e-02
## 1268 -1.741266e-02 6.364251e-02
## 1269 -1.839585e-02 7.602549e-02
## 1270 -6.887158e-02 3.470718e-02
## 1271 1.732636e-02 -2.412421e-02
## 1272 -2.291497e-01 -2.382751e-01
## 1273 -3.628490e-01 -4.345692e-01
## 1274 -1.222667e-01 -5.722811e-02
## 1275 1.699987e-01 -4.478670e-02
## 1276 -1.096994e-01 -5.925302e-02
## 1277 2.208824e-02 7.573833e-02
## 1278 -2.027537e-03 8.147042e-02
## 1279 2.357667e-03 7.915237e-02
## 1280 -2.024796e-02 6.752395e-02
## 1281 -1.188621e-02 8.892574e-02
## 1282 3.215952e-01 -1.676749e-01
## 1283 -1.091569e-02 8.741967e-02
## 1284 -3.182030e-03 8.913989e-02
## 1285 -1.526384e-02 8.148489e-02
## 1286 -2.018182e-02 6.445906e-02
## 1287 -2.676332e-02 5.719728e-02
## 1288 -7.577752e-02 5.181091e-03
## 1289 -2.082843e-02 7.122249e-02
## 1290 -1.654858e-02 6.761078e-02
## 1291 -3.642764e-02 4.042604e-02
## 1292 2.486137e-01 -1.008015e-01
## 1293 4.995021e-03 6.202247e-02
## 1294 -9.093939e-02 -3.428136e-02
## 1295 -2.299968e-01 -2.420129e-01
## 1296 -1.612549e-02 7.869490e-02
## 1297 -1.683805e-02 6.784553e-02
## 1298 -2.599128e-02 5.942698e-02
## 1299 1.273986e-02 1.100067e-01
## 1300 -1.858819e-02 6.681568e-02
## 1301 6.967584e-02 5.168490e-02
## 1302 -2.165269e-02 4.590507e-02
## 1303 2.395751e-01 -6.371106e-02
## 1304 4.150866e-02 3.359106e-02
## 1305 -3.400140e-02 5.073530e-02
## 1306 -1.534226e-01 -1.487662e-01
## 1307 -6.496089e-02 5.761423e-03
## 1308 1.007011e-01 -3.808068e-02
## 1309 1.548873e-03 9.197634e-02
## 1310 -6.035351e-02 2.741526e-02
## 1311 -2.668613e-02 4.102915e-02
## 1312 -1.312418e-02 5.262286e-02
## 1313 -1.852398e-01 -1.822380e-01
## 1314 1.644313e-01 3.076292e-02
## 1315 -3.259150e-01 -3.815624e-01
## 1316 -2.723026e-01 -3.144584e-01
## 1317 -2.772976e-02 4.684833e-02
## 1318 -1.490731e-02 7.018155e-02
## 1319 -1.668690e-01 -1.474678e-01
## 1320 -1.008570e-01 -5.074862e-02
## 1321 -1.433026e-02 6.385696e-02
## 1322 2.643741e-01 -1.108165e-01
## 1323 8.061869e-02 -2.197360e-02
## 1324 2.609611e-01 -1.533314e-01
## 1325 1.505447e-01 -5.714241e-02
## 1326 1.757863e-01 -5.782009e-02
## 1327 2.203803e-01 -1.026414e-01
## 1328 -2.416957e-01 -2.732107e-01
## 1329 3.672341e-01 -1.941018e-01
## 1330 -1.774472e-02 6.348451e-02
## 1331 -6.546431e-02 -3.731765e-03
## 1332 -2.448466e-02 6.667573e-02
## 1333 -2.491975e-01 -2.722786e-01
## 1334 -2.410089e-02 5.567095e-02
## 1335 -5.387265e-02 3.004767e-02
## 1336 -7.848849e-02 9.255586e-03
## 1337 -4.120571e-02 4.958143e-02
## 1338 -2.027422e-02 6.340820e-02
## 1339 6.635897e-03 9.135022e-02
## 1340 -9.889259e-03 9.089125e-02
## 1341 4.956816e-02 7.683768e-02
## 1342 1.844942e-01 -8.662394e-02
## 1343 -1.211978e-02 8.910029e-02
## 1344 -1.184358e-01 -8.782473e-02
## 1345 4.042411e-03 9.593774e-02
## 1346 -1.821050e-01 -1.774003e-01
## 1347 -6.311631e-03 8.012719e-02
## 1348 -8.042518e-03 8.418649e-02
## 1349 -2.824148e-01 -3.216670e-01
## 1350 -2.548500e-02 5.810039e-02
## 1351 -1.781507e-02 6.813901e-02
## 1352 -2.007510e-01 -2.020794e-01
## 1353 5.173493e-02 7.964132e-02
## 1354 4.002036e-02 1.870145e-02
## 1355 1.241474e-01 -5.725630e-02
## 1356 -2.406673e-02 6.588015e-02
## 1357 -2.774735e-01 -3.193928e-01
## 1358 -8.192134e-03 7.276044e-02
## 1359 -2.843496e-01 -3.359038e-01
## 1360 -6.019946e-02 1.391481e-02
## 1361 -1.156218e-02 7.071693e-02
## 1362 -1.731312e-01 -1.613121e-01
## 1363 3.275027e-02 1.123501e-01
## 1364 -2.465019e-03 8.221320e-02
## 1365 -1.425739e-02 7.779683e-02
## 1366 -8.055664e-02 -2.939782e-02
## 1367 3.216667e-02 9.090600e-02
## 1368 -1.908530e-01 -1.926424e-01
## 1369 1.298084e-01 4.778353e-02
## 1370 -3.041391e-02 5.502359e-02
## 1371 -2.265933e-02 4.993884e-02
## 1372 5.095418e-02 5.218967e-03
## 1373 -8.058616e-02 -3.116325e-02
## 1374 -2.636433e-02 6.362987e-02
## 1375 2.906363e-01 -1.800762e-01
## 1376 3.450210e-03 1.084354e-01
## 1377 1.495329e-01 -5.730906e-02
## 1378 -1.867790e-02 6.776242e-02
## 1379 -2.892881e-02 5.034498e-02
## 1380 -1.589964e-02 8.108346e-02
## 1381 -3.251112e-02 5.895684e-02
## 1382 -1.283815e-02 6.656616e-02
## 1383 -7.295826e-02 -1.775235e-02
## 1384 -1.788720e-02 5.443124e-02
## 1385 8.138789e-02 -5.155385e-03
## 1386 1.523027e-01 1.468932e-02
## 1387 -2.834456e-02 5.494061e-02
## 1388 -1.765204e-01 -1.441217e-01
## 1389 -8.503737e-03 5.434576e-02
## 1390 -5.930185e-02 1.549501e-02
## 1391 2.114055e-01 -1.172628e-01
## 1392 -2.719279e-01 -3.089050e-01
## 1393 -1.293798e-01 -1.056890e-01
## 1394 3.323393e-01 -1.634033e-01
## 1395 -3.005538e-02 3.982921e-02
## 1396 -1.204923e-01 -7.594711e-02
## 1397 2.418420e-01 -1.141902e-01
## 1398 1.402841e-02 6.694270e-02
## 1399 -1.516757e-02 6.766701e-02
## 1400 -3.414825e-02 4.774859e-02
## 1401 -2.674625e-02 6.057302e-02
## 1402 2.744042e-01 -1.565974e-01
## 1403 -1.606414e-02 6.410707e-02
## 1404 -5.409990e-02 2.105998e-02
## 1405 2.309074e-02 8.369319e-02
## 1406 -3.185163e-02 6.102805e-02
## 1407 1.134128e-02 4.547275e-02
## 1408 -1.540867e-01 -1.351431e-01
## 1409 5.625092e-02 4.491535e-02
## 1410 3.064274e-01 -1.490560e-01
## 1411 -1.014546e-02 6.456143e-02
## 1412 -3.956586e-02 2.997243e-02
## 1413 -2.864275e-02 5.085985e-02
## 1414 -4.864818e-02 4.384956e-03
## 1415 -1.570747e-02 7.507154e-02
## 1416 2.003280e-02 6.385548e-02
## 1417 -6.939591e-03 1.000525e-01
## 1418 -2.812184e-01 -3.053311e-01
## 1419 3.586106e-01 -1.835743e-01
## 1420 -5.106471e-02 3.062087e-02
## 1421 8.291676e-02 3.860775e-02
## 1422 3.057697e-03 1.067937e-01
## 1423 -1.175839e-01 -7.466063e-02
## 1424 1.228720e-01 5.001106e-02
## 1425 -8.053883e-02 3.115419e-02
## 1426 -1.660839e-01 -1.342816e-01
## 1427 -3.114393e-03 8.573642e-02
## 1428 1.517352e-01 -1.705021e-02
## 1429 -1.638348e-01 -1.042179e-01
## 1430 -5.151520e-02 4.788048e-02
## 1431 1.313365e-01 1.353006e-02
## 1432 4.192603e-01 -2.469819e-01
## 1433 -2.601819e-01 -2.815400e-01
## 1434 1.626623e-02 6.532916e-02
## 1435 1.083800e-01 -2.420486e-02
## 1436 -2.959770e-02 5.985606e-02
## 1437 -1.212609e-01 -7.121917e-02
## 1438 -2.028423e-01 -2.206768e-01
## 1439 -8.460039e-02 -2.081769e-02
## 1440 -2.783416e-01 -3.238234e-01
## 1441 -1.388201e-01 -5.755167e-02
## 1442 -3.560643e-02 4.994232e-02
## 1443 -6.992433e-02 2.020520e-02
## 1444 3.019779e-01 -1.762046e-01
## 1445 -3.960333e-02 4.026388e-02
## 1446 -8.542110e-02 -3.969742e-02
## 1447 -1.841867e-02 5.773112e-02
## 1448 4.615382e-02 9.652376e-02
## 1449 3.898281e-01 -2.178989e-01
## 1450 -2.280919e-01 -2.410577e-01
## 1451 -2.608900e-02 6.428377e-02
## 1452 1.937214e-01 -8.791876e-02
## 1453 -9.063821e-02 -4.225344e-02
## 1454 -5.854633e-02 3.101557e-02
## 1455 -2.203267e-02 5.778575e-02
## 1456 3.493697e-02 9.889625e-02
## 1457 -3.745986e-02 4.263277e-02
## 1458 1.516149e-01 -8.370706e-02
## 1459 -3.127120e-01 -3.771425e-01
## 1460 -1.800741e-01 -1.848890e-01
## 1461 -1.227321e-02 8.908554e-02
## 1462 -1.302964e-01 -9.450271e-02
## 1463 -6.669207e-02 -2.364776e-03
## 1464 -4.664285e-02 4.074829e-02
## 1465 -2.144816e-01 -2.110102e-01
## 1466 -8.343128e-02 -1.016792e-02
## 1467 -1.047960e-02 7.295490e-02
## 1468 -3.119195e-02 5.995745e-02
## 1469 -3.386773e-03 7.133151e-02
## 1470 -6.279062e-02 1.657180e-02
## 1471 -2.442223e-02 6.591411e-02
## 1472 -1.205190e-01 -7.842977e-02
## 1473 5.039375e-02 5.124622e-02
## 1474 -7.973449e-02 -3.254220e-02
## 1475 -4.456551e-02 4.710687e-02
## 1476 -1.215702e-01 -9.902489e-02
## 1477 -1.582532e-01 -1.606685e-01
## 1478 -1.483859e-02 6.466257e-02
## 1479 -9.216925e-02 -4.981745e-02
## 1480 -1.520600e-01 -1.395324e-01
## 1481 -1.010955e-01 -3.642569e-02
## 1482 -2.096845e-01 -2.087200e-01
## 1483 -2.414902e-01 -2.544477e-01
## 1484 -2.287423e-01 -2.555161e-01
## 1485 -2.538952e-02 7.383314e-02
## 1486 -1.936571e-01 -1.963940e-01
## 1487 -4.250535e-02 2.433330e-02
## 1488 -3.218549e-02 5.937850e-02
## 1489 -2.566344e-01 -2.888500e-01
## 1490 1.205315e-03 7.934157e-02
## 1491 -2.295171e-02 6.497767e-02
## 1492 3.476407e-03 1.063478e-01
## 1493 -1.440817e-02 8.696366e-02
## 1494 -3.195122e-02 4.349889e-02
## 1495 -1.176974e-02 8.940357e-02
## 1496 -1.020867e-02 8.535214e-02
## 1497 1.031961e-02 1.044712e-01
## 1498 -9.186932e-02 -2.945933e-02
## 1499 -1.502959e-02 8.821867e-02
## 1500 2.222834e-03 8.404692e-02
## 1501 1.085504e-01 -2.811101e-02
## 1502 -3.988702e-02 2.788614e-02
## 1503 -1.756063e-02 7.683453e-02
## 1504 -2.264415e-01 -2.463592e-01
## 1505 2.763876e-01 -1.439271e-01
## 1506 -1.419697e-02 6.430093e-02
## 1507 -2.461337e-02 7.888816e-02
## 1508 -2.632957e-02 4.492061e-02
## 1509 -3.707272e-02 4.533645e-02
## 1510 -1.012475e-02 7.557532e-02
## 1511 7.180061e-03 3.973907e-02
## 1512 -1.567073e-01 -1.417943e-01
## 1513 -5.701464e-02 3.427055e-02
## 1514 -4.749338e-02 3.982312e-02
## 1515 -2.367590e-02 6.864683e-02
## 1516 -4.272039e-02 3.709598e-02
## 1517 -1.718275e-01 -1.602163e-01
## 1518 -1.040263e-02 8.700434e-02
## 1519 -5.928222e-02 2.414974e-02
## 1520 -4.327824e-02 5.870657e-02
## 1521 3.516354e-01 -2.057041e-01
## 1522 7.460409e-02 7.178265e-02
## 1523 -7.380033e-02 -2.085451e-02
## 1524 -8.560381e-02 -3.952593e-02
## 1525 -3.260842e-02 4.251148e-02
## 1526 -1.677667e-02 6.707998e-02
## 1527 7.735144e-02 1.888934e-02
## 1528 -3.759523e-02 3.826368e-02
## 1529 1.636915e-02 1.540044e-02
## 1530 -2.773039e-02 3.463435e-02
## 1531 7.361602e-02 4.387085e-02
## 1532 -1.227201e-01 -8.562864e-02
## 1533 -8.119785e-03 6.871431e-02
## 1534 5.810885e-03 7.821181e-02
## 1535 -7.383859e-02 4.496504e-03
## 1536 5.507333e-02 3.426889e-02
## 1537 -9.286829e-02 -5.226981e-02
## 1538 -3.129951e-01 -3.687905e-01
## 1539 1.966917e-01 -1.004750e-01
## 1540 4.045844e-01 -1.943286e-01
## 1541 -1.760383e-01 -1.176504e-01
## 1542 1.977140e-01 -9.728184e-02
## 1543 -9.112229e-03 6.904352e-02
## 1544 8.732897e-02 4.796637e-02
## 1545 -1.786460e-02 6.816194e-02
## 1546 1.068589e-01 -9.219018e-03
## 1547 -1.709721e-02 5.994920e-02
## 1548 -9.546910e-03 7.153683e-02
## 1549 3.044286e-01 -1.319590e-01
## 1550 -2.507643e-02 6.980790e-02
## 1551 -2.848844e-02 5.863263e-02
## 1552 2.611551e-01 -1.175078e-01
## 1553 -2.211836e-01 -2.327609e-01
## 1554 -6.248005e-03 7.201546e-02
## 1555 -1.939653e-01 -2.045942e-01
## 1556 2.502266e-02 9.134331e-02
## 1557 8.114156e-02 4.646069e-02
## 1558 -3.048199e-01 -3.550116e-01
## 1559 -4.118402e-02 2.850878e-02
## 1560 -2.279685e-02 7.334446e-02
## 1561 -1.036511e-01 -4.679483e-02
## 1562 2.690989e-02 9.256448e-02
## 1563 -5.778014e-02 1.983220e-02
## 1564 -4.948050e-02 3.352346e-02
## 1565 -1.088303e-02 9.744872e-02
## 1566 -6.248246e-02 5.959470e-03
## 1567 -1.410409e-02 1.077322e-01
## 1568 1.208789e-01 2.007356e-02
## 1569 -9.060317e-03 8.759425e-02
## 1570 1.001467e-01 -1.055253e-02
## 1571 -7.523037e-02 -1.397782e-02
## 1572 1.766525e-02 9.238185e-02
## 1573 -9.170413e-03 9.645459e-02
## 1574 -1.729268e-02 7.220221e-02
## 1575 -9.234090e-02 -5.270957e-02
## 1576 3.298021e-01 -2.193674e-01
## 1577 -1.937058e-02 6.589756e-02
## 1578 -1.193816e-01 -8.200874e-02
## 1579 6.927282e-02 3.215468e-02
## 1580 -4.176886e-02 5.990831e-02
## 1581 -2.309130e-02 6.751119e-02
## 1582 3.411711e-02 9.824588e-02
## 1583 -8.754505e-02 -3.396668e-02
## 1584 1.156036e-01 -3.979046e-03
## 1585 -1.570031e-01 -1.447818e-01
## 1586 7.345600e-02 1.216247e-02
## 1587 3.120558e-02 5.560274e-02
## 1588 -7.232316e-02 3.951557e-02
## 1589 -1.998741e-01 -1.789115e-01
## 1590 -3.319626e-01 -4.050515e-01
## 1591 -9.424739e-03 8.771179e-02
## 1592 6.211435e-02 9.919579e-02
## 1593 -1.998787e-01 -1.759388e-01
## 1594 -1.353584e-01 -1.207567e-01
## 1595 -1.572623e-01 -1.394352e-01
## 1596 -2.768093e-02 3.256235e-02
## 1597 4.733913e-02 6.756831e-02
## 1598 -3.175869e-02 5.364394e-02
## 1599 -1.274779e-02 4.379332e-02
## 1600 -6.633627e-02 -1.087260e-03
## 1601 -9.939480e-03 9.040806e-02
## 1602 5.902613e-02 9.786659e-02
## 1603 -1.716803e-01 -1.678306e-01
## 1604 3.069688e-02 1.158968e-01
## 1605 2.987588e-01 -1.577687e-01
## 1606 2.090181e-01 -9.876138e-02
## 1607 2.873418e-02 3.612946e-02
## 1608 8.931386e-03 1.113597e-01
## 1609 -3.247701e-02 -6.185833e-03
## 1610 -2.767028e-02 4.900033e-02
## 1611 -2.149708e-02 5.646814e-02
## 1612 -4.682356e-02 6.429506e-02
## 1613 -1.975142e-02 6.980916e-02
## 1614 -2.108574e-01 -2.239547e-01
## 1615 4.140156e-01 -2.377891e-01
## 1616 -3.064853e-01 -3.446950e-01
## 1617 -3.770971e-02 4.730679e-02
## 1618 -7.022699e-02 1.958720e-03
## 1619 2.728362e-01 -1.637521e-01
## 1620 -3.441605e-02 4.603593e-02
## 1621 -6.181158e-02 8.009162e-03
## 1622 -3.782539e-02 3.943937e-02
## 1623 -2.116170e-01 -2.347610e-01
## 1624 -1.077503e-01 -7.062502e-02
## 1625 8.069378e-02 6.748847e-02
## 1626 -5.190908e-02 4.271923e-02
## 1627 -8.959725e-02 -3.986012e-02
## 1628 1.740315e-01 6.320975e-04
## 1629 -6.102101e-02 -1.008815e-02
## 1630 -1.574919e-02 6.766310e-02
## 1631 -1.378716e-01 -1.193222e-01
## 1632 -1.530345e-01 -1.427608e-01
## 1633 -5.430386e-03 5.815606e-02
## 1634 -1.121259e-02 1.062399e-01
## 1635 -9.592099e-02 -3.275866e-02
## 1636 -7.629468e-03 1.044014e-01
## 1637 -1.067929e-02 8.388384e-02
## 1638 2.850549e-02 7.565298e-02
## 1639 -3.221717e-01 -3.859722e-01
## 1640 -1.837411e-02 6.921609e-02
## 1641 1.829252e-02 9.594420e-02
## 1642 -6.257934e-03 1.175421e-01
## 1643 -7.572820e-02 5.939848e-03
## 1644 -3.785232e-02 6.568035e-02
## 1645 -1.917501e-02 5.525228e-02
## 1646 -2.708196e-02 5.924209e-02
## 1647 9.782255e-02 -2.651608e-02
## 1648 -1.821678e-02 5.611950e-02
## 1649 -1.290104e-03 7.439098e-02
## 1650 1.742447e-01 -6.527440e-02
## 1651 1.627808e-01 -5.283360e-02
## 1652 3.438012e-01 -1.969545e-01
## 1653 -3.002015e-02 3.272259e-02
## 1654 -2.293056e-02 7.161201e-02
## 1655 1.974146e-02 1.134269e-01
## 1656 3.292246e-01 -1.851691e-01
## 1657 -5.781403e-02 -3.446403e-03
## 1658 -1.325940e-01 -1.040546e-01
## 1659 -1.050105e-02 5.439322e-02
## 1660 -7.543184e-02 -1.046680e-02
## 1661 1.046103e-01 -2.610760e-02
## 1662 -4.385149e-02 3.977387e-02
## 1663 -9.352238e-03 9.141430e-02
## 1664 -9.551956e-02 -1.442252e-02
## 1665 -1.691687e-02 6.492764e-02
## 1666 -2.929003e-02 5.458844e-02
## 1667 4.411196e-02 1.068888e-01
## 1668 -8.206032e-02 -3.212132e-02
## 1669 -4.687486e-02 3.849974e-02
## 1670 -1.765939e-02 7.225424e-02
## 1671 3.147455e-01 -1.553012e-01
## 1672 -2.958199e-01 -3.499997e-01
## 1673 -1.685400e-02 6.171461e-02
## 1674 3.375838e-01 -1.371840e-01
## 1675 5.561663e-02 8.164651e-02
## 1676 -2.171171e-03 5.434510e-02
## 1677 -2.185195e-01 -2.239968e-01
## 1678 -7.949190e-02 -6.944215e-03
## 1679 2.179038e-04 4.542747e-02
## 1680 3.081499e-01 -1.775907e-01
## 1681 -2.769273e-01 -3.291808e-01
## 1682 -2.687632e-02 6.027160e-02
## 1683 2.575371e-01 -1.109474e-01
## 1684 -1.332484e-02 7.764803e-02
## 1685 -3.954156e-02 4.746443e-02
## 1686 -1.122037e-02 9.594893e-02
## 1687 -4.688006e-02 2.354402e-02
## 1688 -1.943813e-01 -1.904713e-01
## 1689 -6.082181e-02 2.782893e-02
## 1690 2.808759e-01 -7.055566e-02
## 1691 -2.201629e-01 -2.389232e-01
## 1692 1.751619e-03 1.250024e-01
## 1693 3.295502e-01 -1.284753e-01
## 1694 -3.709172e-02 4.761755e-02
## 1695 -4.960725e-02 3.728299e-02
## 1696 -1.350191e-01 -8.982980e-02
## 1697 -4.444852e-02 6.172133e-02
## 1698 -1.047450e-01 -4.856788e-02
## 1699 2.491948e-01 -9.395050e-02
## 1700 -1.102692e-01 -4.662547e-02
## 1701 -1.275510e-02 7.385307e-02
## 1702 3.941537e-01 -2.184583e-01
## 1703 1.121692e-02 1.157247e-01
## 1704 -1.549531e-01 -1.498130e-01
## 1705 -1.486807e-01 -1.218813e-01
## 1706 -1.397269e-02 8.796256e-02
## 1707 -4.911632e-03 1.033605e-01
## 1708 -1.092546e-01 -8.780833e-02
## 1709 4.558055e-03 8.697278e-02
## 1710 -8.421627e-02 -3.599306e-02
## 1711 -6.739913e-02 6.085782e-03
## 1712 -4.023498e-02 4.613427e-02
## 1713 2.464878e-01 -6.270126e-02
## 1714 -2.206788e-01 -2.335101e-01
## 1715 3.292966e-01 -2.043042e-01
## 1716 -6.437845e-03 9.566081e-02
## 1717 -1.426189e-02 9.636652e-02
## 1718 -5.559783e-02 1.887410e-02
## 1719 -2.641078e-01 -3.127946e-01
## 1720 -2.834834e-02 5.960803e-02
## 1721 -2.593982e-01 -2.865931e-01
## 1722 3.399637e-01 -1.922362e-01
## 1723 -9.332667e-02 -5.736200e-02
## 1724 -1.671005e-01 -1.624311e-01
## 1725 -1.233333e-02 9.252070e-02
## 1726 -5.951371e-02 8.021299e-03
## 1727 -3.514998e-02 5.784534e-02
## 1728 -6.835641e-03 9.472342e-02
## 1729 7.877330e-02 -2.708817e-03
## 1730 -1.402688e-02 6.662557e-02
## 1731 4.211818e-02 9.284083e-02
## 1732 1.735750e-02 6.542226e-02
## 1733 -5.098672e-03 1.059140e-01
## 1734 -1.891372e-02 6.957942e-02
## 1735 -2.377345e-02 5.267162e-02
## 1736 -2.386680e-02 4.865660e-02
## 1737 -5.791168e-02 3.349494e-02
## 1738 -2.517777e-02 6.290073e-02
## 1739 -1.099732e-02 8.311384e-02
## 1740 1.619078e-01 4.547272e-03
## 1741 3.709709e-01 -1.894751e-01
## 1742 3.371542e-01 -2.098101e-01
## 1743 1.806982e-02 1.125950e-01
## 1744 9.027718e-02 -5.385820e-03
## 1745 -5.465768e-02 2.731527e-03
## 1746 -5.237903e-02 4.029785e-03
## 1747 3.471878e-03 7.172029e-02
## 1748 -7.661959e-02 -7.084012e-04
## 1749 -1.870728e-02 6.627284e-02
## 1750 -1.819241e-01 -1.836372e-01
## 1751 -3.706114e-02 4.243342e-02
## 1752 -1.914991e-02 6.268942e-02
## 1753 -1.969746e-01 -1.993680e-01
## 1754 -8.233288e-02 -1.072454e-02
## 1755 -1.948307e-01 -2.043665e-01
## 1756 -1.660793e-01 -1.640292e-01
## 1757 -1.552776e-02 6.968701e-02
## 1758 -6.030744e-02 6.089099e-03
## 1759 -3.475002e-02 4.213863e-02
## 1760 -4.081097e-02 4.205394e-02
## 1761 7.655059e-03 9.995931e-02
## 1762 -2.898339e-01 -3.483736e-01
## 1763 -2.400701e-01 -2.625734e-01
## 1764 -2.558040e-02 6.035220e-02
## 1765 -1.794725e-02 6.579181e-02
## 1766 -1.654869e-02 6.310815e-02
## 1767 -1.500792e-02 5.386101e-02
## 1768 2.450340e-03 1.140822e-01
## 1769 -1.431122e-02 6.961270e-02
## 1770 -3.272388e-02 5.065053e-02
## 1771 -1.837582e-02 6.557008e-02
## 1772 3.694245e-01 -2.039375e-01
## 1773 -8.797374e-03 7.893493e-02
## 1774 -1.807122e-01 -1.853375e-01
## 1775 4.012084e-02 1.021926e-01
## 1776 1.094432e-01 -2.591597e-02
## 1777 2.868714e-01 -1.036040e-01
## 1778 2.800663e-02 1.381647e-01
## 1779 1.824004e-03 7.148841e-02
## 1780 -1.178351e-01 -6.418413e-02
## 1781 3.673839e-01 -2.254619e-01
## 1782 -3.033179e-02 5.036330e-02
## 1783 -2.613095e-02 5.440757e-02
## 1784 -3.027487e-02 9.161290e-04
## 1785 -1.620897e-02 6.631074e-02
## 1786 -1.562372e-01 -1.502148e-01
## 1787 -2.516234e-01 -2.843041e-01
## 1788 -4.650256e-02 1.460677e-02
## 1789 -2.883350e-02 6.359348e-02
## 1790 3.323128e-01 -1.921676e-01
## 1791 -1.869869e-01 -1.941142e-01
## 1792 -2.117234e-01 -2.203391e-01
## 1793 -1.175327e-01 -9.468572e-02
## 1794 -4.591855e-02 2.602411e-02
## 1795 1.577667e-01 -5.845561e-02
## 1796 3.054757e-01 -1.575151e-01
## 1797 -2.091097e-01 -2.046185e-01
## 1798 -1.695358e-01 -1.485509e-01
## 1799 3.192768e-03 1.060386e-01
## 1800 1.591186e-01 2.152684e-02
## 1801 1.474450e-01 1.892879e-03
## 1802 -1.211483e-01 -9.218901e-02
## 1803 -3.194396e-02 4.747426e-02
## 1804 3.009395e-03 1.173875e-01
## 1805 -1.826209e-01 -1.889867e-01
## 1806 -1.103869e-01 -6.146382e-02
## 1807 -3.176718e-03 8.643052e-02
## 1808 -1.231364e-02 8.958235e-02
## 1809 -2.051137e-02 7.046074e-02
## 1810 -3.185262e-02 5.179857e-02
## 1811 -5.669312e-03 7.486682e-02
## 1812 -9.892915e-03 1.018191e-01
## 1813 -6.508972e-02 1.103587e-03
## 1814 -1.839187e-02 5.457885e-02
## 1815 3.629786e-03 1.192355e-01
## 1816 -9.170522e-03 1.236761e-01
## 1817 -1.147030e-02 8.347787e-02
## 1818 1.112019e-01 1.341234e-02
## 1819 2.750784e-02 7.341370e-02
## 1820 -6.941822e-03 9.960994e-02
## 1821 -9.858916e-02 -2.971253e-02
## 1822 3.484501e-02 7.556571e-02
## 1823 6.299948e-02 6.700446e-02
## 1824 -2.799877e-02 5.373488e-02
## 1825 -3.573804e-02 4.424152e-02
## 1826 -1.816732e-02 6.582265e-02
## 1827 -2.004049e-02 5.413120e-02
## 1828 -2.287648e-03 9.681929e-02
## 1829 -2.166493e-02 6.998086e-02
## 1830 2.608920e-01 -1.390669e-01
## 1831 3.613396e-01 -2.124471e-01
## 1832 -2.512409e-02 4.660041e-02
## 1833 -4.751183e-02 3.697304e-02
## 1834 2.231024e-02 5.892375e-02
## 1835 1.053260e-01 -3.083026e-02
## 1836 -8.255399e-03 6.460383e-02
## 1837 2.321300e-02 1.062751e-01
## 1838 8.482056e-03 7.648071e-02
## 1839 4.269544e-02 7.158113e-02
## 1840 -7.088398e-02 -2.048306e-02
## 1841 2.822705e-01 -1.397232e-01
## 1842 -8.689224e-03 9.484853e-02
## 1843 -4.452114e-04 6.107592e-02
## 1844 -7.100361e-04 2.645934e-02
## 1845 6.855072e-02 5.776646e-02
## 1846 -5.360017e-02 4.763867e-02
## 1847 -9.691169e-02 -1.647833e-02
## 1848 -2.553908e-01 -2.740299e-01
## 1849 -1.131455e-02 6.483580e-02
## 1850 1.731106e-01 -1.096650e-01
## 1851 -4.819867e-02 1.706331e-02
## 1852 3.434613e-01 -1.895497e-01
## 1853 1.238380e-01 6.157292e-02
## 1854 -1.594672e-01 -1.353548e-01
## 1855 -4.931100e-02 5.459357e-02
## 1856 -7.432241e-03 9.027591e-02
## 1857 -2.511172e-02 6.050388e-02
## 1858 -8.738237e-03 9.744919e-02
## 1859 5.468467e-02 6.320518e-02
## 1860 -1.249361e-01 -8.152530e-02
## 1861 4.226370e-03 1.015143e-01
## 1862 -3.215634e-02 5.240916e-02
## 1863 2.723083e-01 -1.085062e-01
## 1864 7.239776e-03 5.612357e-02
## 1865 -6.238905e-03 1.136370e-01
## 1866 -1.928699e-02 8.544547e-02
## 1867 -2.412255e-01 -2.634160e-01
## 1868 -9.800544e-02 -6.485314e-02
## 1869 -3.269615e-02 5.297411e-02
## 1870 -1.799862e-02 6.910617e-02
## 1871 -9.908670e-03 6.101676e-02
## 1872 -4.722385e-02 3.963964e-02
## 1873 2.570620e-03 1.317253e-01
## 1874 -1.850437e-02 6.554825e-02
## 1875 3.499151e-02 7.113140e-02
## 1876 1.885527e-02 1.064614e-01
## 1877 -2.311990e-02 7.915963e-02
## 1878 -2.262506e-02 6.491230e-02
## 1879 -2.863397e-02 5.888444e-02
## 1880 -1.533069e-02 6.661298e-02
## 1881 -8.422329e-02 -4.324513e-02
## 1882 -1.161120e-02 8.444611e-02
## 1883 -1.553371e-02 4.961881e-02
## 1884 -1.372999e-02 5.838210e-02
## 1885 -2.415756e-02 9.051411e-02
## 1886 -1.455249e-02 7.261730e-02
## 1887 3.300360e-01 -2.040143e-01
## 1888 -2.329606e-02 5.628460e-02
## 1889 -1.330472e-02 7.834149e-02
## 1890 -9.533627e-02 -3.458844e-02
## 1891 2.590575e-02 1.031017e-01
## 1892 -2.876205e-02 4.784556e-02
## 1893 2.645211e-01 -1.234118e-01
## 1894 -5.381873e-02 3.294737e-03
## 1895 2.088840e-01 -9.634250e-02
## 1896 3.130218e-02 1.090198e-01
## 1897 1.308829e-01 -1.661047e-03
## 1898 -8.284942e-02 7.879970e-03
## 1899 -2.446000e-02 8.083093e-02
## 1900 -3.466910e-02 4.909343e-02
## 1901 -5.148891e-02 1.714516e-02
## 1902 4.876569e-03 9.672064e-02
## 1903 -2.693637e-02 5.377883e-02
## 1904 -2.526163e-01 -2.926654e-01
## 1905 1.751167e-02 9.314073e-02
## 1906 -1.140258e-02 8.429597e-02
## 1907 3.906706e-02 1.300472e-01
## 1908 -7.023892e-03 4.761876e-02
## 1909 -1.963038e-01 -1.999521e-01
## 1910 1.773652e-01 -2.094370e-03
## 1911 -2.213863e-02 6.743038e-02
## 1912 8.406897e-03 1.159286e-01
## 1913 -1.511829e-01 -1.364040e-01
## 1914 -2.168591e-02 7.318188e-02
## 1915 6.406499e-02 5.764358e-02
## 1916 -2.629276e-02 6.640640e-02
## 1917 -1.705710e-02 6.397236e-02
## 1918 -1.568495e-02 7.237473e-02
## 1919 2.331552e-02 4.038672e-02
## 1920 -6.127099e-02 2.444578e-02
## 1921 -1.451527e-02 8.298664e-02
## 1922 -1.219026e-02 6.519458e-02
## 1923 -2.501666e-01 -2.829737e-01
## 1924 3.160352e-01 -1.867906e-01
## 1925 1.989757e-01 -9.734232e-02
## 1926 -1.991150e-02 7.725804e-02
## 1927 -1.060655e-02 8.460871e-02
## 1928 3.530184e-01 -1.862714e-01
## 1929 5.280107e-02 2.616283e-02
## 1930 -1.587897e-02 5.271124e-02
## 1931 -4.749947e-02 1.416062e-02
## 1932 -4.331579e-02 1.079889e-02
## 1933 5.643886e-02 8.303221e-02
## 1934 -2.312854e-02 6.133135e-02
## 1935 -2.102089e-02 5.931694e-02
## 1936 -1.063372e-02 5.848625e-02
## 1937 -4.334647e-02 2.038880e-02
## 1938 1.785304e-03 1.099692e-01
## 1939 -2.632854e-01 -3.043446e-01
## 1940 -1.541799e-02 6.327208e-02
## 1941 -9.113336e-02 -1.126257e-02
## 1942 -4.126781e-02 6.096404e-02
## 1943 -1.093326e-01 -4.430625e-02
## 1944 -5.080812e-02 3.582337e-02
## 1945 -3.354061e-02 4.446277e-02
## 1946 -1.538694e-02 6.429946e-02
## 1947 2.688315e-01 -1.099596e-01
## 1948 -5.038809e-02 2.579561e-02
## 1949 -4.644269e-02 3.411026e-02
## 1950 -4.365772e-02 2.668715e-02
## 1951 -9.928066e-02 -5.587261e-02
## 1952 -2.237700e-01 -2.336024e-01
## 1953 -1.146787e-01 -8.953962e-02
## 1954 1.404082e-02 9.870896e-02
## 1955 -2.128210e-02 6.062864e-02
## 1956 6.410739e-02 5.630640e-02
## 1957 -1.428641e-02 7.786611e-02
## 1958 -5.451271e-02 3.911565e-02
## 1959 -3.230346e-01 -3.857670e-01
## 1960 -1.402510e-02 8.311625e-02
## 1961 -2.945214e-02 6.145501e-02
## 1962 -2.792531e-02 7.186642e-02
## 1963 8.715774e-02 7.392983e-02
## 1964 -2.690973e-01 -3.110013e-01
## 1965 -1.074799e-01 -6.001354e-02
## 1966 1.288658e-02 9.898317e-02
## 1967 -5.596445e-03 8.615667e-02
## 1968 -2.848181e-02 7.337783e-02
## 1969 -1.005002e-02 6.784546e-02
## 1970 -2.260047e-02 6.914163e-02
## 1971 -2.528316e-01 -2.911926e-01
## 1972 1.752294e-01 -7.288747e-02
## 1973 -1.071641e-01 -7.926528e-02
## 1974 -2.242943e-01 -2.377120e-01
## 1975 -1.238538e-02 1.034662e-01
## 1976 -1.695398e-02 5.506862e-02
## 1977 4.692432e-01 -2.933351e-01
## 1978 4.113794e-03 1.070034e-01
## 1979 -2.676042e-02 5.529361e-02
## 1980 3.246088e-02 1.208516e-01
## 1981 -1.460099e-02 7.331513e-02
## 1982 -1.611037e-02 6.620608e-02
## 1983 -1.997686e-02 6.870382e-02
## 1984 4.249285e-01 -2.393424e-01
## 1985 -5.671034e-03 9.888151e-02
## 1986 -2.012600e-01 -1.960181e-01
## 1987 -8.785395e-03 8.271041e-02
## 1988 -7.107216e-02 1.527129e-02
## 1989 -3.874549e-02 3.544873e-02
## 1990 -1.643142e-02 7.347472e-02
## 1991 -9.314826e-03 9.408563e-02
## 1992 -1.253236e-01 -9.489760e-02
## 1993 -1.000605e-02 9.691073e-02
## 1994 -2.156626e-03 1.212555e-01
## 1995 -7.640765e-04 9.812354e-02
## 1996 -1.479938e-02 7.732712e-02
## 1997 -1.416231e-02 8.823934e-02
## 1998 -1.638666e-01 -1.382663e-01
## 1999 -1.143473e-02 1.084534e-01
## 2000 -8.444635e-02 -4.055696e-02
## 2001 -1.918121e-02 5.368645e-02
## 2002 -1.480626e-02 8.243253e-02
## 2003 -1.581737e-02 6.248004e-02
## 2004 -3.517233e-02 5.805052e-02
## 2005 -6.113304e-02 7.274143e-03
## 2006 -9.329589e-02 -3.848785e-02
## 2007 -3.865503e-02 6.531002e-02
## 2008 -1.304871e-01 -8.575745e-02
## 2009 -9.036943e-02 -5.056954e-02
## 2010 -1.884575e-02 6.863022e-02
## 2011 -2.015044e-02 5.505615e-02
## 2012 4.971911e-03 7.565641e-02
## 2013 3.955790e-02 1.133454e-01
## 2014 -7.989055e-03 5.790577e-02
## 2015 1.273450e-01 4.011147e-02
## 2016 -5.915864e-03 5.435457e-02
## 2017 -3.368626e-03 1.245091e-01
## 2018 -1.392000e-02 8.460360e-02
## 2019 -3.577500e-02 4.867015e-02
## 2020 5.935816e-02 9.328903e-02
## 2021 -3.058423e-02 6.039176e-02
## 2022 -4.003440e-02 2.604997e-02
## 2023 2.755632e-02 6.971797e-02
## 2024 2.775186e-02 1.124838e-01
## 2025 -8.587324e-02 -3.401182e-02
## 2026 -1.942937e-02 6.455239e-02
## 2027 -1.686199e-01 -1.557750e-01
## 2028 -1.440632e-02 6.236764e-02
## 2029 -3.021205e-02 3.944674e-02
## 2030 -1.369840e-02 9.159185e-02
## 2031 -2.521994e-02 8.557331e-02
## 2032 -2.218401e-02 8.508415e-02
## 2033 -2.737512e-02 4.802400e-02
## 2034 -1.496705e-02 4.642442e-02
## 2035 -4.664324e-02 3.478988e-02
## 2036 -1.030303e-01 -5.030660e-02
## 2037 -1.504150e-02 8.808677e-02
## 2038 -2.545650e-01 -2.873902e-01
## 2039 1.788203e-02 8.527034e-02
## 2040 -9.299452e-02 -3.706952e-02
## 2041 -1.865321e-02 6.266270e-02
## 2042 2.211301e-03 1.346370e-01
## 2043 -7.088094e-03 8.837041e-02
## 2044 -2.380045e-02 4.958414e-02
## 2045 3.660896e-01 -1.968661e-01
## 2046 -6.554261e-02 -1.050120e-02
## 2047 1.582502e-01 1.796759e-02
## 2048 -1.579571e-02 5.550314e-02
## 2049 3.924213e-01 -2.390044e-01
## 2050 2.510322e-01 -1.182411e-01
## 2051 2.813951e-01 -1.352250e-01
## 2052 -2.376909e-02 7.919347e-02
## 2053 2.849219e-02 1.069108e-01
## 2054 -4.205297e-02 4.543896e-02
## 2055 2.941368e-01 -1.549779e-01
## 2056 -5.293679e-02 1.522966e-02
## 2057 -1.129842e-02 7.609997e-02
## 2058 1.066939e-02 6.769474e-02
## 2059 -1.756360e-02 6.339048e-02
## 2060 -8.211892e-02 -3.935827e-02
## 2061 -1.101895e-01 -7.763817e-02
## 2062 7.123643e-02 1.531247e-02
## 2063 4.180273e-01 -2.487862e-01
## 2064 4.511608e-03 8.080422e-02
## 2065 -1.458827e-02 6.602309e-02
## 2066 -1.275262e-02 7.743311e-02
## 2067 -1.578650e-01 -1.245800e-01
## 2068 -2.323553e-01 -2.431213e-01
## 2069 -1.245349e-02 7.040415e-02
## 2070 -6.751169e-03 1.166428e-01
## 2071 -8.950535e-02 -4.330524e-02
## 2072 -4.577419e-02 1.550306e-02
## 2073 -8.867507e-02 -3.882061e-02
## 2074 1.447637e-02 8.870336e-02
## 2075 -1.593826e-02 7.579125e-02
## 2076 -2.151848e-02 5.974270e-02
## 2077 2.155822e-02 1.130195e-01
## 2078 -1.387613e-02 1.015990e-01
## 2079 -2.517931e-02 6.899820e-02
## 2080 -1.062942e-01 -7.091402e-02
## 2081 -1.482199e-02 5.904100e-02
## 2082 -3.489400e-02 6.390993e-02
## 2083 -1.057827e-01 -7.035054e-02
## 2084 -7.806180e-02 -2.171659e-02
## 2085 -2.941868e-02 6.069775e-02
## 2086 -3.464625e-02 3.517197e-02
## 2087 -1.626171e-02 6.911009e-02
## 2088 -1.011277e-02 5.542601e-02
## 2089 -1.394793e-02 8.487384e-02
## 2090 -4.844716e-02 2.402386e-02
## 2091 -7.654107e-03 6.999592e-02
## 2092 2.780151e-01 -1.196128e-01
## 2093 -3.057303e-02 6.900268e-02
## 2094 -3.243923e-02 6.526682e-02
## 2095 -1.799735e-01 -1.579346e-01
## 2096 -9.499701e-02 -4.936947e-02
## 2097 -6.238301e-03 1.244786e-01
## 2098 8.313457e-03 1.219603e-01
## 2099 -2.171531e-02 5.733488e-02
## 2100 -7.229242e-03 9.598698e-02
## 2101 -8.326072e-02 -3.573660e-02
## 2102 3.523425e-01 -2.157581e-01
## 2103 3.506178e-02 1.366769e-01
## 2104 4.031310e-03 8.156723e-02
## 2105 -1.505856e-02 6.392829e-02
## 2106 -3.392035e-02 5.223445e-02
## 2107 2.615341e-01 -1.584458e-01
## 2108 4.457471e-02 6.712652e-02
## 2109 6.517262e-02 8.656674e-02
## 2110 -1.836283e-02 4.134345e-02
## 2111 -6.809760e-03 6.485630e-02
## 2112 9.231903e-03 6.192878e-02
## 2113 -6.931320e-02 -6.822341e-03
## 2114 6.714018e-02 1.009354e-02
## 2115 -1.485270e-02 7.110729e-02
## 2116 -3.223203e-02 6.200215e-02
## 2117 -1.414756e-02 6.320738e-02
## 2118 -1.529751e-03 8.704345e-02
## 2119 -4.552224e-02 2.279407e-02
## 2120 1.596545e-02 9.589646e-02
## 2121 -2.902372e-02 4.334023e-02
## 2122 5.895136e-03 9.468154e-02
## 2123 -8.497182e-03 8.958695e-02
## 2124 -4.714147e-02 3.000219e-02
## 2125 -1.729965e-02 7.986900e-02
## 2126 6.242224e-04 9.587838e-02
## 2127 -6.481304e-03 9.307471e-02
## 2128 -1.385547e-02 8.098566e-02
## 2129 -1.263601e-02 1.057369e-01
## 2130 -1.622887e-02 1.046480e-01
## 2131 8.077060e-02 9.678314e-02
## 2132 -2.511038e-02 8.668009e-02
## 2133 -3.179970e-03 1.167936e-01
## 2134 -1.100139e-02 9.049487e-02
## 2135 -8.394731e-02 -2.175673e-02
## 2136 -2.769325e-02 8.165806e-02
## 2137 -1.775593e-02 7.117108e-02
## 2138 -8.680782e-02 -3.654537e-02
## 2139 -4.370575e-02 2.222784e-02
## 2140 -3.645719e-02 7.267276e-02
## 2141 2.222159e-02 4.077999e-02
## 2142 -2.227365e-02 6.323823e-02
## 2143 -3.389302e-02 4.673573e-02
## 2144 -1.266332e-02 7.443542e-02
## 2145 1.512130e-02 1.258065e-01
## 2146 -1.620964e-02 6.940740e-02
## 2147 -8.654923e-03 9.269386e-02
## 2148 -2.028616e-02 7.530408e-02
## 2149 4.383023e-02 9.656030e-02
## 2150 -1.682837e-02 6.720105e-02
## 2151 -1.640370e-02 6.521451e-02
## 2152 -2.299393e-02 4.373673e-02
## 2153 -2.319491e-02 6.925203e-02
## 2154 -7.614368e-02 -1.284482e-02
## 2155 -5.775777e-03 1.032783e-01
## 2156 1.499841e-01 -4.602970e-04
## 2157 7.521272e-02 8.566415e-02
## 2158 -2.723061e-02 9.699210e-02
## 2159 -8.184251e-03 1.298439e-01
## 2160 -1.151973e-02 7.360453e-02
## 2161 -2.226707e-02 8.649290e-02
## 2162 -7.224423e-03 7.683922e-02
## 2163 6.460392e-03 1.474687e-01
## 2164 2.002763e-01 -1.011570e-01
## 2165 8.045537e-02 1.911425e-02
## 2166 3.831451e-01 -2.050254e-01
## 2167 9.399907e-02 5.382238e-02
## 2168 -6.686636e-03 6.698688e-02
## 2169 -4.685045e-02 3.522842e-02
## 2170 8.943300e-02 1.048902e-01
## 2171 -1.668543e-02 5.461291e-02
## 2172 -3.443241e-02 3.072424e-02
## 2173 3.748977e-01 -2.021282e-01
## 2174 -3.155197e-03 1.041752e-01
## 2175 -5.366454e-02 4.775492e-03
## 2176 -7.323560e-03 6.313916e-02
## 2177 -1.735084e-02 7.336150e-02
## 2178 -1.381466e-01 -1.043570e-01
## 2179 -5.074197e-03 6.760882e-02
## 2180 2.285187e-01 -2.585886e-02
## 2181 -2.793385e-02 5.117596e-02
## 2182 -6.892430e-03 8.032846e-02
## 2183 -4.080386e-02 4.182129e-02
## 2184 -2.069071e-03 1.368944e-01
## 2185 -2.111617e-02 6.487596e-02
## 2186 -9.141742e-02 -4.509822e-02
## 2187 -1.532192e-02 8.231495e-02
## 2188 -5.478028e-02 8.261332e-03
## 2189 -1.480957e-01 -1.171142e-01
## 2190 6.027663e-02 7.207010e-02
## 2191 -1.033862e-01 -6.257750e-02
## 2192 -1.494904e-02 7.820472e-02
## 2193 -2.182548e-02 5.582416e-02
## 2194 -2.231500e-02 6.470177e-02
## 2195 -1.098884e-02 1.004773e-01
## 2196 -1.612121e-02 7.397897e-02
## 2197 -9.988443e-03 7.825201e-02
## 2198 -5.140576e-02 2.284640e-02
## 2199 -2.653946e-02 8.724228e-02
## 2200 -2.098672e-02 6.442556e-02
## 2201 -2.239621e-02 8.240210e-02
## 2202 -1.387690e-02 8.492138e-02
## 2203 -9.395016e-02 -3.714400e-02
## 2204 -1.739777e-02 8.568992e-02
## 2205 -3.829005e-02 5.688476e-02
## 2206 5.595303e-02 7.498557e-02
## 2207 -6.994574e-03 8.796140e-02
## 2208 -9.460981e-02 -3.655581e-02
## 2209 -1.798234e-02 6.624885e-02
## 2210 8.578337e-02 2.861638e-02
## 2211 -3.487410e-02 4.550329e-02
## 2212 3.990292e-02 7.232258e-02
## 2213 -1.216275e-02 7.548964e-02
## 2214 -1.166023e-02 1.050410e-01
## 2215 -3.307946e-02 4.869167e-02
## 2216 -1.466874e-02 1.065296e-01
## 2217 6.623648e-02 9.370337e-02
## 2218 -3.192098e-02 6.038007e-02
## 2219 -1.722061e-02 7.270768e-02
## 2220 -2.685828e-02 6.828420e-02
## 2221 -9.446610e-02 -5.339897e-02
## 2222 -1.237813e-02 7.018027e-02
## 2223 -1.872114e-02 6.520223e-02
## 2224 1.778763e-01 -8.043752e-02
## 2225 3.946906e-02 3.882507e-02
## 2226 -4.810955e-02 1.269468e-02
## 2227 -5.368053e-02 2.787338e-02
## 2228 -2.258827e-02 7.098675e-02
## 2229 -4.718324e-02 3.273909e-02
## 2230 -9.639642e-03 6.676389e-02
## 2231 -2.642756e-02 6.705297e-02
## 2232 1.034779e-01 5.633608e-02
## 2233 3.219177e-01 -1.601036e-01
## 2234 2.763572e-01 -1.470323e-01
## 2235 -1.446343e-02 6.375719e-02
## 2236 -1.015023e-02 5.298989e-02
## 2237 8.597950e-02 8.916929e-02
## 2238 2.367397e-02 9.377095e-02
## 2239 4.250327e-03 1.198558e-01
## 2240 -5.325099e-03 8.480936e-02
## 2241 -2.096233e-02 6.910305e-02
## 2242 -1.243998e-02 7.159861e-02
## 2243 1.791187e-02 1.003685e-01
## 2244 -2.884425e-02 4.723687e-02
## 2245 -2.913409e-02 4.938684e-02
## 2246 3.156780e-01 -2.031192e-01
## 2247 -2.879012e-02 4.710814e-02
## 2248 3.390407e-03 1.264725e-01
## 2249 -2.004604e-02 6.868793e-02
## 2250 -1.588086e-02 6.973558e-02
## 2251 -2.694759e-02 6.187021e-02
## 2252 -1.862087e-02 5.471494e-02
## 2253 -2.942536e-02 7.547327e-02
## 2254 -1.016503e-01 -4.640957e-02
## 2255 -2.403527e-02 5.514086e-02
## 2256 -1.548176e-02 7.672292e-02
## 2257 3.322196e-01 -1.684918e-01
## 2258 -7.116857e-03 1.250715e-01
## 2259 -1.329058e-02 7.900019e-02
## 2260 -2.304467e-03 1.146564e-01
## 2261 -9.165971e-03 8.257596e-02
## 2262 -1.448514e-02 8.263664e-02
## 2263 3.038833e-01 -1.447207e-01
## 2264 -2.157954e-02 8.010489e-02
## 2265 -5.953854e-03 1.112161e-01
## 2266 -7.436679e-03 1.034209e-01
## 2267 -4.660214e-02 2.798493e-02
## 2268 -1.718537e-02 6.787230e-02
## 2269 -3.150978e-02 1.051731e-01
## 2270 -1.878440e-02 6.879236e-02
## 2271 -1.844688e-02 6.174129e-02
## 2272 7.658440e-02 9.283279e-02
## 2273 2.116083e-02 4.446814e-02
## 2274 -1.573844e-02 8.115358e-02
## 2275 2.682144e-02 1.268825e-01
## 2276 -2.980414e-02 5.400815e-02
## 2277 3.627110e-02 6.629557e-02
## 2278 -1.441836e-02 7.870004e-02
## 2279 1.748335e-02 1.272216e-01
## 2280 3.484989e-01 -1.696919e-01
## 2281 -1.240944e-01 -9.335290e-02
## 2282 3.374473e-01 -1.705943e-01
## 2283 -6.996518e-03 9.460775e-02
## 2284 -1.185454e-02 7.324412e-02
## 2285 -1.117990e-02 7.450271e-02
## 2286 -4.462316e-02 1.804615e-02
## 2287 4.576449e-01 -2.576385e-01
## 2288 5.389075e-03 8.003008e-02
## 2289 -8.832811e-03 7.184604e-02
## 2290 1.109094e-04 1.028338e-01
## 2291 -2.521295e-02 5.487935e-02
## 2292 3.668892e-01 -1.946628e-01
## 2293 -5.135119e-03 1.040570e-01
## 2294 1.092338e-02 3.734946e-02
## 2295 7.761361e-02 8.232059e-02
## 2296 7.173461e-02 1.091070e-01
## 2297 -8.202773e-02 -2.103107e-02
## 2298 2.970141e-02 8.983895e-02
## 2299 -1.937251e-02 6.229543e-02
## 2300 3.259078e-01 -1.691688e-01
## 2301 -3.272236e-02 3.644730e-02
## 2302 -1.109292e-01 -5.497774e-02
## 2303 5.591998e-03 9.893809e-02
## 2304 -3.607561e-02 4.081181e-02
## 2305 8.936105e-02 7.706428e-02
## 2306 9.782343e-03 1.077110e-01
## 2307 -2.431655e-02 4.287499e-02
## 2308 1.375657e-02 9.921633e-02
## 2309 -1.135593e-01 -6.420669e-02
## 2310 -7.723531e-02 -2.571545e-02
## 2311 -9.479585e-02 -3.815416e-02
## 2312 -3.275644e-02 5.031806e-02
## 2313 -2.056131e-02 5.988811e-02
## 2314 -5.036247e-02 3.572462e-02
## 2315 -2.039913e-02 8.350822e-02
## 2316 -1.732667e-02 6.885107e-02
## 2317 -9.525566e-03 1.133400e-01
## 2318 -1.452269e-02 6.654317e-02
## 2319 -7.261429e-02 -8.336823e-03
## 2320 -8.184596e-02 -3.526492e-03
## 2321 -1.793614e-02 7.291831e-02
## 2322 -1.284395e-02 7.163548e-02
## 2323 -9.504684e-02 -3.792820e-02
## 2324 -1.351215e-02 6.820163e-02
## 2325 -2.320138e-02 7.948251e-02
## 2326 -2.052274e-02 8.241223e-02
## 2327 2.523135e-01 -9.073705e-02
## 2328 -9.092327e-03 5.879989e-02
## 2329 -3.447404e-02 3.745080e-02
## 2330 -1.592514e-02 7.238394e-02
## 2331 -2.576601e-02 4.461678e-02
## 2332 -2.069534e-02 7.886897e-02
## 2333 -1.486036e-02 1.003996e-01
## 2334 -8.201581e-03 7.988482e-02
## 2335 -7.899247e-03 7.861294e-02
## 2336 -1.297778e-02 7.726137e-02
## 2337 -2.001352e-02 6.742956e-02
## 2338 -6.624789e-02 1.825899e-03
## 2339 -7.216722e-03 1.178298e-01
## 2340 1.188727e-01 5.533894e-02
## 2341 1.578716e-02 4.590657e-02
## 2342 -5.303502e-02 7.201219e-03
## 2343 3.652706e-03 6.070858e-02
## 2344 -1.245357e-02 5.934897e-02
## 2345 -1.860842e-02 7.343515e-02
## 2346 -4.569880e-02 2.352473e-02
## 2347 4.670870e-02 1.100305e-01
## 2348 4.329957e-01 -2.475366e-01
## 2349 -1.700360e-02 7.822251e-02
## 2350 -1.109567e-02 1.084105e-01
## 2351 -8.646322e-04 1.060086e-01
## 2352 -6.825278e-03 5.894880e-02
## 2353 3.064207e-02 4.483195e-02
## 2354 3.689004e-01 -2.103507e-01
## 2355 -7.653536e-02 -2.843673e-02
## 2356 -3.147115e-02 7.262969e-02
## 2357 1.611390e-01 4.204964e-02
## 2358 -2.793665e-02 5.317835e-02
## 2359 -5.424742e-03 1.130490e-01
## 2360 4.707475e-04 4.846898e-02
## 2361 6.465663e-02 1.278908e-02
## 2362 -2.629998e-02 6.271015e-02
## 2363 1.285732e-01 6.147946e-02
## 2364 -1.019082e-01 -6.351688e-02
## 2365 4.348791e-02 8.988930e-02
## 2366 -6.625216e-02 1.446016e-04
## 2367 -1.586065e-02 8.473992e-02
## 2368 -6.173361e-02 7.716942e-03
## 2369 -1.631857e-02 7.833058e-02
## 2370 5.600198e-02 7.432192e-02
## 2371 -9.484289e-03 6.614548e-02
## 2372 4.216567e-01 -2.354409e-01
## 2373 -1.943241e-02 7.577628e-02
## 2374 -1.924445e-02 5.674765e-02
## 2375 -4.663810e-02 1.574203e-02
## 2376 -7.699053e-02 -2.228507e-02
## 2377 -1.819655e-02 7.511733e-02
## 2378 -2.230489e-02 5.839258e-02
## 2379 -6.903617e-02 4.736630e-03
## 2380 -2.544123e-02 8.316254e-02
## 2381 7.693832e-03 9.989054e-02
## 2382 6.377498e-02 -4.427342e-03
## 2383 -1.855582e-02 6.504849e-02
## 2384 -1.972411e-02 6.330393e-02
## 2385 -4.074421e-03 5.511486e-02
## 2386 -2.287437e-02 5.670420e-02
## 2387 -6.959364e-03 1.041048e-01
## 2388 -1.535269e-02 6.828032e-02
## 2389 -1.759696e-02 6.292212e-02
## 2390 -9.664264e-02 -5.313514e-02
## 2391 -4.945613e-02 1.775596e-02
## 2392 1.576130e-03 8.708248e-02
## 2393 -1.938359e-02 5.643768e-02
## 2394 -8.815839e-02 -3.349115e-02
## 2395 -6.482025e-03 1.099643e-01
## 2396 2.829235e-01 -1.280775e-01
## 2397 -1.363603e-02 7.793401e-02
## 2398 -1.026644e-02 6.867268e-02
## 2399 2.879114e-02 8.257634e-02
## 2400 -1.520215e-02 5.792473e-02
## 2401 -3.730714e-02 5.730840e-02
## 2402 -2.265991e-02 9.217766e-02
## 2403 -2.849321e-02 7.327958e-02
## 2404 -2.734178e-02 7.111550e-02
## 2405 -6.123410e-02 -6.539145e-03
## 2406 -1.991812e-02 7.837283e-02
## 2407 -8.003617e-03 7.811188e-02
## 2408 -1.645330e-02 7.209976e-02
## 2409 -2.221229e-02 7.666790e-02
## 2410 3.585157e-02 9.650719e-02
## 2411 -2.622410e-02 9.214843e-02
## 2412 1.718763e-02 8.042434e-02
## 2413 -2.042072e-02 6.529392e-02
## 2414 -2.763495e-02 6.202781e-02
## 2415 -2.574823e-02 5.345225e-02
## 2416 3.073702e-02 5.825670e-02
## 2417 2.445267e-02 5.579517e-02
## 2418 -4.527369e-02 2.194408e-02
## 2419 -1.064637e-02 8.337897e-02
## 2420 -1.727067e-02 6.335840e-02
## 2421 2.632080e-01 -5.714823e-02
## 2422 1.981805e-02 9.523663e-02
## 2423 6.082865e-02 9.228920e-02
## 2424 -3.803448e-03 9.406398e-02
## 2425 -1.487916e-02 8.968482e-02
## 2426 -2.598837e-02 8.186795e-02
## 2427 -4.987599e-02 1.343095e-02
## 2428 -3.981325e-02 4.435177e-02
## 2429 -4.635434e-03 1.156539e-01
## 2430 -1.209406e-03 1.002339e-01
## 2431 -1.838512e-02 6.179891e-02
## 2432 -3.336011e-02 4.505273e-02
## 2433 -1.581389e-02 8.211473e-02
## 2434 -2.526648e-02 5.629026e-02
## 2435 -4.139327e-02 3.454505e-02
## 2436 -6.467876e-02 1.188101e-02
## 2437 -2.503806e-02 5.251851e-02
## 2438 -1.628077e-02 7.008115e-02
## 2439 -2.549323e-02 7.172259e-02
## 2440 -2.047513e-02 8.088962e-02
## 2441 -1.487052e-02 6.140137e-02
## 2442 -1.623775e-02 7.502970e-02
## 2443 -1.289892e-02 6.796841e-02
## 2444 -4.657283e-02 3.231595e-02
## 2445 -1.120981e-02 7.815495e-02
## 2446 -1.629105e-02 1.040089e-01
## 2447 6.827174e-02 5.574199e-02
## 2448 -2.574694e-02 9.624403e-02
## 2449 9.313490e-02 -7.817911e-03
## 2450 -1.851017e-02 8.405967e-02
## 2451 -2.243440e-02 8.577223e-02
## 2452 -1.219702e-02 7.899308e-02
## 2453 2.768572e-03 9.585647e-02
## 2454 -1.574146e-02 6.965401e-02
## 2455 -1.675299e-02 6.800297e-02
## 2456 -5.798962e-02 4.876540e-03
## 2457 -1.531170e-02 8.933803e-02
## 2458 -1.064660e-02 1.042287e-01
## 2459 -2.073037e-02 7.837039e-02
## 2460 -1.676890e-02 7.938343e-02
## 2461 -1.502124e-02 7.427676e-02
## 2462 -2.015270e-02 8.417109e-02
## 2463 1.083298e-01 -3.226056e-02
## 2464 -8.619471e-03 9.237059e-02
## 2465 -7.881898e-03 8.699506e-02
## 2466 2.639405e-02 9.418103e-02
## 2467 -4.171353e-03 1.007600e-01
## 2468 -1.655044e-02 8.050252e-02
## 2469 -7.085083e-03 1.000414e-01
## 2470 -5.140291e-03 1.265121e-01
## 2471 -9.041507e-03 9.916057e-02
## 2472 -1.238675e-02 7.383900e-02
## 2473 -9.582155e-03 8.167937e-02
## 2474 -2.775157e-02 5.127376e-02
## 2475 -5.680922e-02 1.867635e-02
## 2476 4.280701e-02 7.064292e-02
## 2477 -4.182308e-02 2.829416e-02
## 2478 -2.168655e-02 8.545504e-02
## 2479 -2.373238e-02 5.583480e-02
## 2480 -2.026077e-02 7.342854e-02
## 2481 -1.393137e-02 8.655232e-02
## 2482 6.591501e-02 -2.781777e-03
## 2483 6.472674e-03 5.479503e-02
## 2484 -1.632709e-02 8.219724e-02
## 2485 -4.533967e-02 2.641904e-02
## 2486 3.656364e-01 -2.133791e-01
## 2487 1.640167e-03 1.205799e-01
## 2488 -1.796167e-02 6.768103e-02
## 2489 -1.579649e-02 7.603015e-02
## 2490 -1.942150e-02 8.266531e-02
## 2491 -3.334929e-02 5.537176e-02
## 2492 -4.607470e-02 3.766712e-02
## 2493 -1.547251e-02 5.971184e-02
## 2494 -1.097564e-02 9.498544e-02
## 2495 2.381106e-01 -1.319978e-01
## 2496 -2.092947e-02 5.055958e-02
## 2497 -1.183829e-02 9.178250e-02
## 2498 1.282811e-03 1.249058e-01
## 2499 -1.438649e-02 7.849231e-02
## 2500 -1.729831e-02 7.863045e-02
## 2501 4.126233e-01 -2.378462e-01
## 2502 -1.585440e-02 7.141747e-02
## 2503 -1.935360e-02 7.280192e-02
## 2504 -2.946424e-02 5.684346e-02
## 2505 -5.009570e-02 2.536252e-02
## 2506 -3.740001e-02 2.775089e-02
## 2507 -1.018800e-02 8.798213e-02
## 2508 -1.795318e-02 6.204994e-02
## 2509 -3.616703e-02 3.700281e-02
## 2510 -2.310908e-02 5.235364e-02
## 2511 6.350948e-02 2.692692e-02
## 2512 4.371299e-01 -2.293118e-01
## 2513 -3.123598e-02 4.571339e-02
## 2514 -2.064692e-02 6.059117e-02
## 2515 9.275567e-02 6.557015e-02
## 2516 -7.983072e-03 1.066657e-01
## 2517 -2.962494e-02 5.579835e-02
## 2518 -1.385806e-03 1.001482e-01
## 2519 -4.743294e-02 4.016855e-02
## 2520 -1.984021e-02 6.153641e-02
## 2521 2.773875e-01 -1.543300e-01
## 2522 -8.464425e-03 8.955463e-02
## 2523 -1.845990e-02 6.911239e-02
## 2524 -1.530341e-02 6.252380e-02
## 2525 -4.034339e-02 4.190072e-02
## 2526 -1.727236e-02 6.195449e-02
## 2527 -2.010199e-02 8.093499e-02
## 2528 -1.719088e-02 7.967719e-02
## 2529 -1.159320e-02 6.750458e-02
## 2530 -2.079465e-02 4.428524e-02
## 2531 -1.175759e-02 9.117831e-02
## 2532 -2.527496e-02 8.084169e-02
## 2533 -2.317550e-02 7.712938e-02
## 2534 -1.244964e-02 8.083009e-02
## 2535 -2.113476e-02 6.806426e-02
## 2536 -1.609317e-02 6.573818e-02
## 2537 -1.938465e-02 6.630264e-02
## 2538 -2.719651e-02 7.062746e-02
## 2539 -2.219018e-02 8.607320e-02
## 2540 -1.975717e-02 7.820925e-02
## 2541 -2.450161e-02 5.576512e-02
## 2542 -6.337605e-02 1.160553e-03
## 2543 -5.348834e-02 2.827797e-02
## 2544 -1.661335e-02 6.838721e-02
## 2545 -2.230698e-02 8.473215e-02
## 2546 -2.063906e-02 6.515709e-02
## 2547 -2.415695e-02 8.594143e-02
## 2548 7.759426e-02 1.091927e-01
## 2549 -4.935317e-02 2.533802e-02
## 2550 1.096768e-01 4.006466e-02
## 2551 -1.499001e-02 7.574651e-02
## 2552 -7.450466e-03 9.349618e-02
## 2553 -8.785823e-02 -3.604009e-02
## 2554 -1.412831e-02 6.862862e-02
## 2555 -8.555333e-02 -3.570022e-02
## 2556 1.251924e-01 4.916965e-02
## 2557 -7.555866e-03 8.826105e-02
## 2558 -1.718793e-02 7.701536e-02
## 2559 1.405605e-02 8.665068e-02
## 2560 -9.711831e-02 -4.796720e-02
## 2561 1.411223e-02 9.259760e-02
## 2562 -1.955081e-02 6.062116e-02
## 2563 -2.067964e-02 5.776991e-02
## 2564 -2.126780e-02 8.465822e-02
## 2565 -1.552526e-02 6.688461e-02
## 2566 -1.085018e-02 1.061512e-01
## 2567 -1.786359e-02 8.468595e-02
## 2568 1.067328e-01 -3.772517e-02
## 2569 1.722367e-02 1.234825e-01
## 2570 -1.577335e-02 6.159030e-02
## 2571 -1.962573e-02 7.879311e-02
## 2572 -2.189885e-02 8.922390e-02
## 2573 -1.755985e-02 5.292921e-02
## 2574 -5.854045e-03 1.006507e-01
## 2575 -2.958497e-02 4.668073e-02
## 2576 -2.965545e-03 6.905526e-02
## 2577 -2.264625e-02 4.964055e-02
## 2578 -1.505236e-02 7.642707e-02
## 2579 -2.019674e-02 6.256269e-02
## 2580 -4.236228e-02 5.267864e-02
## 2581 -1.955458e-02 7.850382e-02
## 2582 -1.445063e-02 6.696492e-02
## 2583 -1.437731e-02 1.047265e-01
## 2584 -2.192619e-02 8.474992e-02
## 2585 -6.689631e-02 6.514266e-04
## 2586 -2.389058e-02 9.165086e-02
## 2587 -1.858873e-02 6.718348e-02
## 2588 -5.043752e-02 2.804625e-02
## 2589 -1.834313e-02 5.120517e-02
## 2590 -1.604436e-02 4.927768e-02
## 2591 1.618302e-03 8.713669e-02
## 2592 -3.395193e-03 4.631729e-02
## 2593 -1.900246e-02 8.025419e-02
## 2594 -1.381187e-02 5.888477e-02
## 2595 -3.908057e-02 2.842485e-02
## 2596 -3.238748e-02 3.724718e-02
## 2597 -1.508061e-02 6.648695e-02
## 2598 5.095909e-02 1.654924e-02
## 2599 9.242236e-02 -1.847173e-02
## 2600 2.770943e-01 -1.492224e-01
## 2601 -1.717151e-02 7.468197e-02
## 2602 -2.098874e-02 6.728764e-02
## 2603 -1.593195e-02 6.090587e-02
## 2604 -8.561677e-02 -2.186658e-02
## 2605 -1.802056e-02 6.319759e-02
## 2606 -1.679599e-02 6.215302e-02
## 2607 -1.930211e-02 6.940245e-02
## 2608 -1.867820e-02 5.696320e-02
## 2609 1.138513e-01 -2.260896e-02
## 2610 -6.844480e-02 8.946632e-04
## 2611 -3.354762e-02 4.285667e-02
## 2612 -1.388750e-02 8.409088e-02
## 2613 -1.939786e-02 8.032807e-02
## 2614 6.181376e-02 3.411990e-02
## 2615 -1.120751e-02 8.121276e-02
## 2616 -1.942308e-02 6.262819e-02
## 2617 -1.545495e-02 8.563498e-02
## 2618 -1.653496e-02 6.546095e-02
## 2619 -1.680460e-02 7.187519e-02
## 2620 -1.911355e-02 7.459205e-02
## 2621 1.179503e-01 -1.416050e-02
## 2622 -1.765192e-02 6.783616e-02
## 2623 -1.709269e-02 7.006766e-02
## 2624 -8.879349e-03 9.583573e-02
## 2625 -1.358159e-02 7.504940e-02
## 2626 -2.146967e-02 6.296901e-02
## 2627 -1.907694e-02 6.529193e-02
## 2628 -1.868409e-02 7.122325e-02
## 2629 -1.659893e-02 5.559075e-02
## 2630 -1.440008e-03 1.003651e-01
## 2631 9.559150e-02 3.718538e-02
## 2632 -2.434837e-02 8.609689e-02
## 2633 1.057849e-01 1.145238e-02
## 2634 -1.812374e-02 6.527840e-02
## 2635 -1.301225e-02 8.810173e-02
## 2636 -1.775780e-02 8.421322e-02
## 2637 -3.448333e-02 3.560492e-02
## 2638 1.155216e-03 1.157451e-01
## 2639 -1.646654e-02 6.044369e-02
## 2640 -1.373327e-02 6.389098e-02
## 2641 5.529506e-02 8.608505e-02
## 2642 -2.327355e-02 7.177264e-02
## 2643 3.678880e-03 5.038934e-02
## 2644 -2.321592e-02 4.936674e-02
## 2645 -8.558403e-03 6.614254e-02
## 2646 -1.459996e-02 6.487822e-02
## 2647 -4.829039e-02 2.067245e-02
## 2648 -1.646608e-02 6.134384e-02
## 2649 -1.168769e-02 6.489649e-02
## 2650 1.576058e-01 -2.270635e-03
## 2651 -2.508836e-02 4.776582e-02
## 2652 1.609746e-02 9.141033e-02
## 2653 -1.825139e-02 6.297245e-02
## 2654 -1.827147e-02 5.673931e-02
## 2655 -2.116849e-02 6.979609e-02
## 2656 -1.545702e-02 7.142045e-02
## 2657 -6.504343e-02 5.376839e-03
## 2658 -1.786944e-02 6.279207e-02
## 2659 1.786264e-01 -6.418291e-02
## 2660 -1.539834e-02 6.646250e-02
## 2661 -2.524136e-02 6.731802e-02
## 2662 -1.444465e-02 7.370711e-02
## 2663 -1.568547e-02 5.567660e-02
## 2664 -1.406381e-02 6.583635e-02
## 2665 -1.609973e-02 6.593410e-02
## 2666 1.416843e-02 8.373666e-02
## 2667 -6.665105e-03 8.216067e-02
##
## $eig
## [1] 3.947033e+01 2.785304e+01 2.113279e+01 1.660350e+01
## [5] 1.514453e+01 1.369256e+01 1.236608e+01 1.200663e+01
## [9] 1.086000e+01 1.054257e+01 1.040641e+01 9.896183e+00
## [13] 9.447034e+00 8.845980e+00 8.400266e+00 8.129663e+00
## [17] 7.846819e+00 7.735503e+00 7.560077e+00 6.881846e+00
## [21] 6.724321e+00 6.505764e+00 6.455684e+00 6.229806e+00
## [25] 6.087511e+00 6.024707e+00 5.891958e+00 5.773067e+00
## [29] 5.705472e+00 5.545264e+00 5.481257e+00 5.337754e+00
## [33] 5.295979e+00 5.223134e+00 5.135956e+00 5.052003e+00
## [37] 4.985266e+00 4.918590e+00 4.861479e+00 4.811359e+00
## [41] 4.725782e+00 4.713585e+00 4.663872e+00 4.571594e+00
## [45] 4.517310e+00 4.485587e+00 4.404087e+00 4.383477e+00
## [49] 4.281294e+00 4.245951e+00 4.197766e+00 4.100257e+00
## [53] 4.097683e+00 4.038224e+00 4.010094e+00 3.979057e+00
## [57] 3.953718e+00 3.918488e+00 3.865423e+00 3.842007e+00
## [61] 3.781778e+00 3.749657e+00 3.727488e+00 3.715595e+00
## [65] 3.694624e+00 3.646652e+00 3.605212e+00 3.588211e+00
## [69] 3.557018e+00 3.539244e+00 3.516298e+00 3.480705e+00
## [73] 3.463639e+00 3.435639e+00 3.414019e+00 3.394163e+00
## [77] 3.377407e+00 3.363110e+00 3.333023e+00 3.312725e+00
## [81] 3.298132e+00 3.283461e+00 3.253585e+00 3.238265e+00
## [85] 3.227048e+00 3.211254e+00 3.197394e+00 3.178268e+00
## [89] 3.157654e+00 3.149127e+00 3.143928e+00 3.120566e+00
## [93] 3.103262e+00 3.079943e+00 3.064265e+00 3.048501e+00
## [97] 3.033122e+00 3.012494e+00 3.009934e+00 2.995965e+00
## [101] 2.976853e+00 2.959604e+00 2.946298e+00 2.932501e+00
## [105] 2.907423e+00 2.903891e+00 2.892829e+00 2.881053e+00
## [109] 2.867748e+00 2.856755e+00 2.835590e+00 2.832441e+00
## [113] 2.826991e+00 2.820342e+00 2.793801e+00 2.791216e+00
## [117] 2.780021e+00 2.761109e+00 2.756851e+00 2.750793e+00
## [121] 2.729070e+00 2.724937e+00 2.709325e+00 2.702045e+00
## [125] 2.694880e+00 2.680996e+00 2.668761e+00 2.653505e+00
## [129] 2.651023e+00 2.643463e+00 2.638347e+00 2.626930e+00
## [133] 2.624792e+00 2.607071e+00 2.601529e+00 2.586097e+00
## [137] 2.584415e+00 2.579095e+00 2.568976e+00 2.559095e+00
## [141] 2.553750e+00 2.544389e+00 2.530808e+00 2.527547e+00
## [145] 2.519760e+00 2.500972e+00 2.496668e+00 2.486945e+00
## [149] 2.481880e+00 2.470168e+00 2.467518e+00 2.460720e+00
## [153] 2.445496e+00 2.432701e+00 2.427539e+00 2.418019e+00
## [157] 2.415739e+00 2.406809e+00 2.405231e+00 2.398677e+00
## [161] 2.391633e+00 2.381092e+00 2.377904e+00 2.367492e+00
## [165] 2.360841e+00 2.357742e+00 2.344557e+00 2.337019e+00
## [169] 2.332420e+00 2.329121e+00 2.320611e+00 2.317781e+00
## [173] 2.308191e+00 2.297816e+00 2.295692e+00 2.287247e+00
## [177] 2.281822e+00 2.278233e+00 2.270881e+00 2.263566e+00
## [181] 2.258764e+00 2.251266e+00 2.249296e+00 2.239760e+00
## [185] 2.237010e+00 2.232133e+00 2.225193e+00 2.223398e+00
## [189] 2.218074e+00 2.207894e+00 2.204665e+00 2.196574e+00
## [193] 2.194958e+00 2.184993e+00 2.181569e+00 2.177559e+00
## [197] 2.172402e+00 2.163906e+00 2.159550e+00 2.156195e+00
## [201] 2.151170e+00 2.145963e+00 2.144030e+00 2.135409e+00
## [205] 2.129365e+00 2.126321e+00 2.116270e+00 2.112545e+00
## [209] 2.108550e+00 2.105241e+00 2.093301e+00 2.088861e+00
## [213] 2.079760e+00 2.076305e+00 2.071209e+00 2.069877e+00
## [217] 2.059898e+00 2.056446e+00 2.055802e+00 2.051879e+00
## [221] 2.046017e+00 2.041215e+00 2.036132e+00 2.031628e+00
## [225] 2.026240e+00 2.018754e+00 2.015952e+00 2.008679e+00
## [229] 2.007374e+00 2.002453e+00 2.000914e+00 1.994058e+00
## [233] 1.988446e+00 1.985601e+00 1.979495e+00 1.977317e+00
## [237] 1.971740e+00 1.965614e+00 1.963217e+00 1.954265e+00
## [241] 1.951471e+00 1.946011e+00 1.942204e+00 1.941587e+00
## [245] 1.935370e+00 1.932993e+00 1.928598e+00 1.923142e+00
## [249] 1.917690e+00 1.913623e+00 1.911162e+00 1.905311e+00
## [253] 1.902506e+00 1.893735e+00 1.890561e+00 1.883801e+00
## [257] 1.883035e+00 1.881639e+00 1.876563e+00 1.874248e+00
## [261] 1.869767e+00 1.861788e+00 1.857630e+00 1.851448e+00
## [265] 1.846106e+00 1.845093e+00 1.842416e+00 1.834325e+00
## [269] 1.833182e+00 1.828979e+00 1.824026e+00 1.822527e+00
## [273] 1.813647e+00 1.810844e+00 1.806880e+00 1.802880e+00
## [277] 1.800927e+00 1.797693e+00 1.790182e+00 1.789162e+00
## [281] 1.784892e+00 1.780304e+00 1.777216e+00 1.773885e+00
## [285] 1.770740e+00 1.766170e+00 1.762172e+00 1.760480e+00
## [289] 1.758333e+00 1.755212e+00 1.752368e+00 1.746861e+00
## [293] 1.740200e+00 1.738129e+00 1.736073e+00 1.733717e+00
## [297] 1.729613e+00 1.725966e+00 1.721026e+00 1.718204e+00
## [301] 1.717526e+00 1.715272e+00 1.708763e+00 1.706419e+00
## [305] 1.704770e+00 1.703184e+00 1.699244e+00 1.698272e+00
## [309] 1.691916e+00 1.686964e+00 1.686240e+00 1.682781e+00
## [313] 1.680072e+00 1.677018e+00 1.673800e+00 1.669652e+00
## [317] 1.666708e+00 1.659857e+00 1.657845e+00 1.655842e+00
## [321] 1.652831e+00 1.648163e+00 1.647293e+00 1.645415e+00
## [325] 1.643300e+00 1.639334e+00 1.635445e+00 1.629095e+00
## [329] 1.626916e+00 1.624668e+00 1.623990e+00 1.617121e+00
## [333] 1.615903e+00 1.613219e+00 1.610599e+00 1.607679e+00
## [337] 1.604730e+00 1.601253e+00 1.598995e+00 1.597645e+00
## [341] 1.591808e+00 1.589047e+00 1.586866e+00 1.579462e+00
## [345] 1.578567e+00 1.577678e+00 1.569722e+00 1.568689e+00
## [349] 1.567292e+00 1.566343e+00 1.562158e+00 1.559632e+00
## [353] 1.554291e+00 1.551818e+00 1.550368e+00 1.547893e+00
## [357] 1.544694e+00 1.543137e+00 1.539882e+00 1.538190e+00
## [361] 1.535749e+00 1.533301e+00 1.532205e+00 1.528349e+00
## [365] 1.525753e+00 1.522876e+00 1.519802e+00 1.516759e+00
## [369] 1.515838e+00 1.513491e+00 1.511347e+00 1.508333e+00
## [373] 1.505749e+00 1.504908e+00 1.499436e+00 1.496475e+00
## [377] 1.494594e+00 1.493210e+00 1.489411e+00 1.488408e+00
## [381] 1.486904e+00 1.484084e+00 1.478474e+00 1.478146e+00
## [385] 1.475174e+00 1.474231e+00 1.471398e+00 1.464607e+00
## [389] 1.462403e+00 1.459625e+00 1.455346e+00 1.452661e+00
## [393] 1.450992e+00 1.449214e+00 1.445842e+00 1.443497e+00
## [397] 1.440489e+00 1.436463e+00 1.435128e+00 1.432640e+00
## [401] 1.431388e+00 1.427231e+00 1.426152e+00 1.423319e+00
## [405] 1.422763e+00 1.420288e+00 1.417269e+00 1.415589e+00
## [409] 1.412448e+00 1.410197e+00 1.409479e+00 1.406273e+00
## [413] 1.403695e+00 1.402680e+00 1.400512e+00 1.399136e+00
## [417] 1.391777e+00 1.389759e+00 1.388599e+00 1.386385e+00
## [421] 1.385928e+00 1.383056e+00 1.381110e+00 1.376345e+00
## [425] 1.375191e+00 1.373573e+00 1.369148e+00 1.367233e+00
## [429] 1.366882e+00 1.364316e+00 1.363782e+00 1.361676e+00
## [433] 1.358630e+00 1.357537e+00 1.352385e+00 1.349290e+00
## [437] 1.348129e+00 1.344479e+00 1.340790e+00 1.340035e+00
## [441] 1.336224e+00 1.335515e+00 1.332756e+00 1.330139e+00
## [445] 1.326888e+00 1.325588e+00 1.324283e+00 1.321186e+00
## [449] 1.320278e+00 1.318906e+00 1.317654e+00 1.315455e+00
## [453] 1.311800e+00 1.310025e+00 1.307052e+00 1.306299e+00
## [457] 1.303514e+00 1.299617e+00 1.298911e+00 1.297191e+00
## [461] 1.296292e+00 1.293093e+00 1.291887e+00 1.288380e+00
## [465] 1.286784e+00 1.284251e+00 1.281423e+00 1.280318e+00
## [469] 1.278162e+00 1.277519e+00 1.274558e+00 1.271320e+00
## [473] 1.269952e+00 1.266856e+00 1.264739e+00 1.261408e+00
## [477] 1.259124e+00 1.257153e+00 1.254698e+00 1.252989e+00
## [481] 1.252183e+00 1.251011e+00 1.248589e+00 1.246172e+00
## [485] 1.244897e+00 1.241375e+00 1.239469e+00 1.237513e+00
## [489] 1.236939e+00 1.233720e+00 1.231685e+00 1.228352e+00
## [493] 1.225949e+00 1.225433e+00 1.224258e+00 1.222120e+00
## [497] 1.220021e+00 1.217195e+00 1.216133e+00 1.213826e+00
## [501] 1.210333e+00 1.209809e+00 1.206346e+00 1.204815e+00
## [505] 1.201867e+00 1.200358e+00 1.198107e+00 1.196082e+00
## [509] 1.194291e+00 1.192639e+00 1.192314e+00 1.189038e+00
## [513] 1.188184e+00 1.183275e+00 1.182320e+00 1.181578e+00
## [517] 1.180143e+00 1.177113e+00 1.175575e+00 1.174035e+00
## [521] 1.170142e+00 1.168709e+00 1.166643e+00 1.166275e+00
## [525] 1.162857e+00 1.162155e+00 1.159322e+00 1.157650e+00
## [529] 1.155707e+00 1.153209e+00 1.151140e+00 1.150083e+00
## [533] 1.148128e+00 1.146207e+00 1.144857e+00 1.143035e+00
## [537] 1.141526e+00 1.140839e+00 1.138408e+00 1.133981e+00
## [541] 1.132916e+00 1.131766e+00 1.130492e+00 1.127197e+00
## [545] 1.126263e+00 1.124339e+00 1.122769e+00 1.121725e+00
## [549] 1.120389e+00 1.117053e+00 1.115118e+00 1.113445e+00
## [553] 1.111528e+00 1.108896e+00 1.107270e+00 1.103244e+00
## [557] 1.101760e+00 1.100627e+00 1.099886e+00 1.096475e+00
## [561] 1.094684e+00 1.094547e+00 1.090673e+00 1.089396e+00
## [565] 1.086712e+00 1.085907e+00 1.084032e+00 1.083072e+00
## [569] 1.082473e+00 1.081585e+00 1.080016e+00 1.078394e+00
## [573] 1.075247e+00 1.074480e+00 1.072234e+00 1.070344e+00
## [577] 1.067400e+00 1.065482e+00 1.064463e+00 1.061542e+00
## [581] 1.059559e+00 1.057564e+00 1.055537e+00 1.054158e+00
## [585] 1.053691e+00 1.051010e+00 1.050275e+00 1.049233e+00
## [589] 1.047610e+00 1.046393e+00 1.043513e+00 1.040224e+00
## [593] 1.039712e+00 1.038679e+00 1.035729e+00 1.034790e+00
## [597] 1.029781e+00 1.029722e+00 1.027936e+00 1.026657e+00
## [601] 1.023562e+00 1.022851e+00 1.021326e+00 1.020373e+00
## [605] 1.017894e+00 1.015347e+00 1.014776e+00 1.012384e+00
## [609] 1.011664e+00 1.009223e+00 1.008332e+00 1.006826e+00
## [613] 1.004777e+00 1.003975e+00 1.002618e+00 9.999259e-01
## [617] 9.987813e-01 9.968799e-01 9.957302e-01 9.917991e-01
## [621] 9.901167e-01 9.897286e-01 9.872210e-01 9.862628e-01
## [625] 9.825057e-01 9.820324e-01 9.802057e-01 9.793369e-01
## [629] 9.781229e-01 9.766386e-01 9.749331e-01 9.739341e-01
## [633] 9.714515e-01 9.703024e-01 9.697488e-01 9.683494e-01
## [637] 9.658512e-01 9.639937e-01 9.608097e-01 9.593784e-01
## [641] 9.584685e-01 9.580188e-01 9.559774e-01 9.540550e-01
## [645] 9.527857e-01 9.512933e-01 9.493536e-01 9.484200e-01
## [649] 9.460498e-01 9.451505e-01 9.438293e-01 9.426074e-01
## [653] 9.407966e-01 9.400523e-01 9.384903e-01 9.370296e-01
## [657] 9.360037e-01 9.341242e-01 9.326513e-01 9.304286e-01
## [661] 9.277458e-01 9.274048e-01 9.238026e-01 9.229483e-01
## [665] 9.215412e-01 9.210678e-01 9.182045e-01 9.165558e-01
## [669] 9.140576e-01 9.132067e-01 9.119856e-01 9.103989e-01
## [673] 9.079915e-01 9.069984e-01 9.064688e-01 9.047207e-01
## [677] 9.036349e-01 9.011058e-01 8.999375e-01 8.991841e-01
## [681] 8.985403e-01 8.952570e-01 8.948372e-01 8.943757e-01
## [685] 8.927746e-01 8.916560e-01 8.888490e-01 8.869541e-01
## [689] 8.864432e-01 8.845835e-01 8.827148e-01 8.818280e-01
## [693] 8.815872e-01 8.790335e-01 8.779287e-01 8.764466e-01
## [697] 8.745321e-01 8.738508e-01 8.726316e-01 8.715487e-01
## [701] 8.700065e-01 8.675551e-01 8.659733e-01 8.647259e-01
## [705] 8.633305e-01 8.626869e-01 8.615206e-01 8.598209e-01
## [709] 8.578720e-01 8.561242e-01 8.539504e-01 8.534969e-01
## [713] 8.519533e-01 8.504954e-01 8.491278e-01 8.477737e-01
## [717] 8.464346e-01 8.434497e-01 8.416888e-01 8.409214e-01
## [721] 8.400039e-01 8.393817e-01 8.381558e-01 8.366280e-01
## [725] 8.337236e-01 8.320302e-01 8.311338e-01 8.280660e-01
## [729] 8.263588e-01 8.251898e-01 8.240955e-01 8.236308e-01
## [733] 8.226645e-01 8.211768e-01 8.178544e-01 8.169510e-01
## [737] 8.153465e-01 8.148931e-01 8.134196e-01 8.115666e-01
## [741] 8.103508e-01 8.099846e-01 8.063535e-01 8.054483e-01
## [745] 8.042333e-01 8.029305e-01 8.018040e-01 7.995917e-01
## [749] 7.992193e-01 7.977856e-01 7.966862e-01 7.953189e-01
## [753] 7.934143e-01 7.928580e-01 7.908221e-01 7.898351e-01
## [757] 7.887053e-01 7.856233e-01 7.846295e-01 7.840404e-01
## [761] 7.817752e-01 7.811906e-01 7.800800e-01 7.787566e-01
## [765] 7.765360e-01 7.742018e-01 7.738785e-01 7.728172e-01
## [769] 7.724427e-01 7.696701e-01 7.693155e-01 7.688748e-01
## [773] 7.657740e-01 7.649060e-01 7.634922e-01 7.613010e-01
## [777] 7.601040e-01 7.596750e-01 7.579714e-01 7.566998e-01
## [781] 7.559059e-01 7.550283e-01 7.529420e-01 7.519746e-01
## [785] 7.503234e-01 7.491111e-01 7.484362e-01 7.478946e-01
## [789] 7.448258e-01 7.428153e-01 7.419537e-01 7.412485e-01
## [793] 7.402815e-01 7.382733e-01 7.364311e-01 7.360464e-01
## [797] 7.355399e-01 7.337238e-01 7.322274e-01 7.311394e-01
## [801] 7.295625e-01 7.281417e-01 7.270037e-01 7.261064e-01
## [805] 7.251767e-01 7.228049e-01 7.216090e-01 7.186537e-01
## [809] 7.168330e-01 7.156283e-01 7.151021e-01 7.148100e-01
## [813] 7.122051e-01 7.119006e-01 7.115209e-01 7.105115e-01
## [817] 7.090241e-01 7.073926e-01 7.061828e-01 7.039466e-01
## [821] 7.015582e-01 7.014086e-01 7.005634e-01 6.991164e-01
## [825] 6.989157e-01 6.966456e-01 6.949728e-01 6.937635e-01
## [829] 6.924408e-01 6.906985e-01 6.899475e-01 6.889181e-01
## [833] 6.867509e-01 6.852423e-01 6.850800e-01 6.825828e-01
## [837] 6.819775e-01 6.804916e-01 6.799175e-01 6.791574e-01
## [841] 6.777446e-01 6.770536e-01 6.760499e-01 6.750627e-01
## [845] 6.732833e-01 6.724078e-01 6.702238e-01 6.693071e-01
## [849] 6.684241e-01 6.665745e-01 6.657617e-01 6.636964e-01
## [853] 6.621298e-01 6.617168e-01 6.606240e-01 6.577912e-01
## [857] 6.571522e-01 6.559206e-01 6.551433e-01 6.550014e-01
## [861] 6.537828e-01 6.511853e-01 6.500290e-01 6.484020e-01
## [865] 6.476458e-01 6.466438e-01 6.449018e-01 6.444518e-01
## [869] 6.425072e-01 6.421874e-01 6.406731e-01 6.377274e-01
## [873] 6.370044e-01 6.364849e-01 6.351853e-01 6.336602e-01
## [877] 6.322749e-01 6.315117e-01 6.299403e-01 6.280036e-01
## [881] 6.273840e-01 6.254214e-01 6.246636e-01 6.234717e-01
## [885] 6.231246e-01 6.207383e-01 6.193749e-01 6.187583e-01
## [889] 6.181879e-01 6.163142e-01 6.160907e-01 6.132992e-01
## [893] 6.129799e-01 6.120664e-01 6.103068e-01 6.081683e-01
## [897] 6.076008e-01 6.065335e-01 6.052248e-01 6.040887e-01
## [901] 6.031362e-01 6.024467e-01 6.005508e-01 5.989041e-01
## [905] 5.977424e-01 5.969932e-01 5.964300e-01 5.959364e-01
## [909] 5.940107e-01 5.930489e-01 5.914963e-01 5.904662e-01
## [913] 5.890068e-01 5.884260e-01 5.866702e-01 5.853885e-01
## [917] 5.844671e-01 5.839246e-01 5.828606e-01 5.820296e-01
## [921] 5.805180e-01 5.793611e-01 5.775217e-01 5.766620e-01
## [925] 5.759930e-01 5.751084e-01 5.746501e-01 5.738800e-01
## [929] 5.715671e-01 5.695162e-01 5.687409e-01 5.683021e-01
## [933] 5.677726e-01 5.669311e-01 5.663721e-01 5.639560e-01
## [937] 5.623565e-01 5.609365e-01 5.600621e-01 5.592828e-01
## [941] 5.578410e-01 5.573904e-01 5.556850e-01 5.551474e-01
## [945] 5.532742e-01 5.520399e-01 5.511679e-01 5.488852e-01
## [949] 5.476847e-01 5.473717e-01 5.458316e-01 5.446615e-01
## [953] 5.441563e-01 5.435886e-01 5.415275e-01 5.405465e-01
## [957] 5.398727e-01 5.368905e-01 5.355189e-01 5.353072e-01
## [961] 5.336032e-01 5.327353e-01 5.316927e-01 5.307335e-01
## [965] 5.290603e-01 5.280322e-01 5.268097e-01 5.265487e-01
## [969] 5.260306e-01 5.241790e-01 5.216547e-01 5.211013e-01
## [973] 5.198500e-01 5.186009e-01 5.171060e-01 5.162979e-01
## [977] 5.156400e-01 5.136135e-01 5.129053e-01 5.123865e-01
## [981] 5.098275e-01 5.080552e-01 5.079169e-01 5.072589e-01
## [985] 5.067496e-01 5.053514e-01 5.040393e-01 5.026376e-01
## [989] 5.020114e-01 5.006398e-01 5.001115e-01 4.987563e-01
## [993] 4.982102e-01 4.967326e-01 4.958024e-01 4.934117e-01
## [997] 4.930954e-01 4.916019e-01 4.911674e-01 4.907156e-01
## [1001] 4.884762e-01 4.876610e-01 4.858385e-01 4.847837e-01
## [1005] 4.836183e-01 4.821529e-01 4.814787e-01 4.810944e-01
## [1009] 4.786216e-01 4.785324e-01 4.771340e-01 4.768795e-01
## [1013] 4.760599e-01 4.746080e-01 4.734185e-01 4.728035e-01
## [1017] 4.716607e-01 4.700514e-01 4.699620e-01 4.679491e-01
## [1021] 4.670977e-01 4.658159e-01 4.650839e-01 4.639216e-01
## [1025] 4.626309e-01 4.614649e-01 4.599555e-01 4.580473e-01
## [1029] 4.578522e-01 4.573616e-01 4.559068e-01 4.552818e-01
## [1033] 4.536246e-01 4.525103e-01 4.515559e-01 4.504637e-01
## [1037] 4.495925e-01 4.485075e-01 4.474795e-01 4.470856e-01
## [1041] 4.465925e-01 4.451087e-01 4.428205e-01 4.421369e-01
## [1045] 4.405869e-01 4.398207e-01 4.386223e-01 4.385603e-01
## [1049] 4.378761e-01 4.373968e-01 4.366048e-01 4.339595e-01
## [1053] 4.333299e-01 4.323250e-01 4.310325e-01 4.309354e-01
## [1057] 4.285384e-01 4.281538e-01 4.270149e-01 4.261829e-01
## [1061] 4.240780e-01 4.226260e-01 4.223091e-01 4.214849e-01
## [1065] 4.199827e-01 4.188693e-01 4.174357e-01 4.168420e-01
## [1069] 4.162238e-01 4.151349e-01 4.144775e-01 4.134486e-01
## [1073] 4.125060e-01 4.115400e-01 4.096996e-01 4.089139e-01
## [1077] 4.084041e-01 4.080689e-01 4.071886e-01 4.051338e-01
## [1081] 4.049484e-01 4.036197e-01 4.024737e-01 4.021270e-01
## [1085] 4.001544e-01 3.998735e-01 3.988334e-01 3.982991e-01
## [1089] 3.975720e-01 3.952299e-01 3.944607e-01 3.936742e-01
## [1093] 3.924696e-01 3.918639e-01 3.910878e-01 3.905198e-01
## [1097] 3.895910e-01 3.882185e-01 3.879040e-01 3.873318e-01
## [1101] 3.865419e-01 3.844262e-01 3.827971e-01 3.818508e-01
## [1105] 3.815459e-01 3.804263e-01 3.797290e-01 3.778410e-01
## [1109] 3.763453e-01 3.748262e-01 3.745590e-01 3.734005e-01
## [1113] 3.730044e-01 3.713161e-01 3.706886e-01 3.682990e-01
## [1117] 3.673443e-01 3.669380e-01 3.663912e-01 3.660947e-01
## [1121] 3.648264e-01 3.636767e-01 3.624869e-01 3.607406e-01
## [1125] 3.584166e-01 3.579781e-01 3.565952e-01 3.561384e-01
## [1129] 3.550919e-01 3.539661e-01 3.523214e-01 3.516654e-01
## [1133] 3.508689e-01 3.496632e-01 3.491597e-01 3.482726e-01
## [1137] 3.473637e-01 3.456404e-01 3.449096e-01 3.443991e-01
## [1141] 3.427271e-01 3.423989e-01 3.418016e-01 3.396571e-01
## [1145] 3.385978e-01 3.375350e-01 3.365062e-01 3.342021e-01
## [1149] 3.339371e-01 3.335188e-01 3.330342e-01 3.320682e-01
## [1153] 3.310830e-01 3.304675e-01 3.302765e-01 3.289298e-01
## [1157] 3.273264e-01 3.272839e-01 3.262216e-01 3.244009e-01
## [1161] 3.231616e-01 3.220943e-01 3.210609e-01 3.200066e-01
## [1165] 3.188631e-01 3.179760e-01 3.172601e-01 3.169918e-01
## [1169] 3.151198e-01 3.140240e-01 3.138737e-01 3.129145e-01
## [1173] 3.124619e-01 3.113552e-01 3.102795e-01 3.095668e-01
## [1177] 3.089831e-01 3.082999e-01 3.062126e-01 3.054419e-01
## [1181] 3.041615e-01 3.039843e-01 3.033676e-01 3.023430e-01
## [1185] 3.004565e-01 2.983886e-01 2.976040e-01 2.967263e-01
## [1189] 2.964885e-01 2.957875e-01 2.946363e-01 2.937733e-01
## [1193] 2.930404e-01 2.920304e-01 2.912268e-01 2.899059e-01
## [1197] 2.889115e-01 2.881624e-01 2.871528e-01 2.859017e-01
## [1201] 2.856598e-01 2.839238e-01 2.821701e-01 2.816965e-01
## [1205] 2.810901e-01 2.803763e-01 2.800413e-01 2.789734e-01
## [1209] 2.779210e-01 2.766719e-01 2.745169e-01 2.739496e-01
## [1213] 2.737604e-01 2.725707e-01 2.717782e-01 2.710730e-01
## [1217] 2.703230e-01 2.689753e-01 2.689364e-01 2.676819e-01
## [1221] 2.666100e-01 2.654794e-01 2.649890e-01 2.637537e-01
## [1225] 2.620433e-01 2.615018e-01 2.609086e-01 2.591229e-01
## [1229] 2.585230e-01 2.578099e-01 2.567541e-01 2.557131e-01
## [1233] 2.551616e-01 2.538789e-01 2.533761e-01 2.507561e-01
## [1237] 2.500999e-01 2.495452e-01 2.491484e-01 2.481058e-01
## [1241] 2.469570e-01 2.463141e-01 2.454776e-01 2.450064e-01
## [1245] 2.444158e-01 2.423361e-01 2.412967e-01 2.403405e-01
## [1249] 2.395703e-01 2.386941e-01 2.383829e-01 2.371193e-01
## [1253] 2.365282e-01 2.355357e-01 2.336962e-01 2.328787e-01
## [1257] 2.315827e-01 2.306909e-01 2.300917e-01 2.294960e-01
## [1261] 2.288149e-01 2.272177e-01 2.260290e-01 2.256677e-01
## [1265] 2.247833e-01 2.242719e-01 2.237551e-01 2.232332e-01
## [1269] 2.215687e-01 2.200130e-01 2.195275e-01 2.194049e-01
## [1273] 2.178256e-01 2.174220e-01 2.159463e-01 2.151868e-01
## [1277] 2.137040e-01 2.126325e-01 2.116409e-01 2.107753e-01
## [1281] 2.102085e-01 2.096221e-01 2.088682e-01 2.086982e-01
## [1285] 2.069568e-01 2.059184e-01 2.058421e-01 2.039737e-01
## [1289] 2.030893e-01 2.030254e-01 2.015040e-01 2.007341e-01
## [1293] 1.988345e-01 1.985683e-01 1.971201e-01 1.956314e-01
## [1297] 1.955677e-01 1.949352e-01 1.940987e-01 1.925923e-01
## [1301] 1.917885e-01 1.906291e-01 1.890493e-01 1.889337e-01
## [1305] 1.881706e-01 1.874987e-01 1.863714e-01 1.859471e-01
## [1309] 1.849835e-01 1.837440e-01 1.823880e-01 1.821247e-01
## [1313] 1.816152e-01 1.803606e-01 1.797152e-01 1.787388e-01
## [1317] 1.782032e-01 1.758232e-01 1.754949e-01 1.742486e-01
## [1321] 1.730597e-01 1.724729e-01 1.722967e-01 1.711633e-01
## [1325] 1.704881e-01 1.691094e-01 1.677707e-01 1.666587e-01
## [1329] 1.662287e-01 1.646221e-01 1.640545e-01 1.632878e-01
## [1333] 1.630307e-01 1.615776e-01 1.602008e-01 1.593602e-01
## [1337] 1.587842e-01 1.576127e-01 1.570318e-01 1.558308e-01
## [1341] 1.553056e-01 1.544207e-01 1.536310e-01 1.532356e-01
## [1345] 1.524192e-01 1.519588e-01 1.508946e-01 1.502777e-01
## [1349] 1.490379e-01 1.470743e-01 1.465671e-01 1.457702e-01
## [1353] 1.453893e-01 1.440984e-01 1.437379e-01 1.422735e-01
## [1357] 1.413315e-01 1.406444e-01 1.400095e-01 1.385854e-01
## [1361] 1.379048e-01 1.372796e-01 1.361358e-01 1.350554e-01
## [1365] 1.345287e-01 1.334900e-01 1.330783e-01 1.320277e-01
## [1369] 1.313733e-01 1.308416e-01 1.292261e-01 1.288141e-01
## [1373] 1.278060e-01 1.269238e-01 1.254617e-01 1.252608e-01
## [1377] 1.246507e-01 1.232218e-01 1.217013e-01 1.214063e-01
## [1381] 1.200538e-01 1.190444e-01 1.183398e-01 1.172134e-01
## [1385] 1.156050e-01 1.147541e-01 1.144292e-01 1.128659e-01
## [1389] 1.124361e-01 1.113008e-01 1.106905e-01 1.102256e-01
## [1393] 1.091058e-01 1.080559e-01 1.070534e-01 1.066772e-01
## [1397] 1.057320e-01 1.042491e-01 1.038259e-01 1.030304e-01
## [1401] 1.020523e-01 1.010558e-01 1.001278e-01 9.999997e-02
## [1405] 9.871498e-02 9.789208e-02 9.708583e-02 9.632919e-02
## [1409] 9.614336e-02 9.554742e-02 9.525529e-02 9.399243e-02
## [1413] 9.205591e-02 9.150988e-02 9.079682e-02 8.995701e-02
## [1417] 8.921291e-02 8.824793e-02 8.746952e-02 8.593175e-02
## [1421] 8.555635e-02 8.532795e-02 8.346277e-02 8.257158e-02
## [1425] 8.212237e-02 8.137034e-02 8.040089e-02 8.030656e-02
## [1429] 7.927977e-02 7.830562e-02 7.771284e-02 7.669069e-02
## [1433] 7.599467e-02 7.530567e-02 7.451922e-02 7.442575e-02
## [1437] 7.250913e-02 7.156089e-02 7.126057e-02 7.003729e-02
## [1441] 6.963760e-02 6.846177e-02 6.732232e-02 6.699991e-02
## [1445] 6.540650e-02 6.507467e-02 6.417920e-02 6.276691e-02
## [1449] 6.189196e-02 6.178276e-02 6.124702e-02 6.042555e-02
## [1453] 6.011628e-02 5.842887e-02 5.811253e-02 5.770996e-02
## [1457] 5.685035e-02 5.588169e-02 5.558905e-02 5.401184e-02
## [1461] 5.312899e-02 5.208132e-02 5.117418e-02 5.104701e-02
## [1465] 5.026862e-02 4.971489e-02 4.921976e-02 4.702415e-02
## [1469] 4.618210e-02 4.587206e-02 4.385873e-02 4.370564e-02
## [1473] 4.320264e-02 4.178571e-02 3.990688e-02 3.929633e-02
## [1477] 3.892000e-02 3.822949e-02 3.772134e-02 3.664748e-02
## [1481] 3.522583e-02 3.479779e-02 3.415034e-02 3.343584e-02
## [1485] 3.277484e-02 3.231955e-02 3.103882e-02 3.047480e-02
## [1489] 2.974159e-02 2.881764e-02 2.810620e-02 2.757020e-02
## [1493] 2.653711e-02 2.614573e-02 2.557157e-02 2.414145e-02
## [1497] 2.375684e-02 2.182326e-02 2.104061e-02 2.016627e-02
## [1501] 1.952740e-02 1.889805e-02 1.813935e-02 1.737330e-02
## [1505] 1.715065e-02 1.556056e-02 1.412035e-02 1.323045e-02
## [1509] 1.255398e-02 1.185841e-02 1.087004e-02 1.070259e-02
## [1513] 8.579872e-03 8.374325e-03 7.159169e-03 6.041210e-03
## [1517] 5.105640e-03 4.336922e-03 3.884967e-03 2.432211e-03
## [1521] 2.199388e-03 1.165057e-03 6.839318e-04 -2.031058e-14
## [1525] -4.230146e-04 -8.119518e-04 -1.398495e-03 -2.799563e-03
## [1529] -3.127780e-03 -4.109185e-03 -4.658540e-03 -5.280037e-03
## [1533] -6.144996e-03 -6.605716e-03 -6.827923e-03 -7.965035e-03
## [1537] -8.981387e-03 -9.318568e-03 -9.661836e-03 -1.167455e-02
## [1541] -1.228408e-02 -1.257863e-02 -1.379968e-02 -1.424007e-02
## [1545] -1.544917e-02 -1.638875e-02 -1.749363e-02 -1.814399e-02
## [1549] -1.891762e-02 -1.958192e-02 -2.077392e-02 -2.190491e-02
## [1553] -2.264474e-02 -2.308602e-02 -2.403635e-02 -2.431966e-02
## [1557] -2.529867e-02 -2.622637e-02 -2.711218e-02 -2.806654e-02
## [1561] -2.887651e-02 -2.965217e-02 -3.008670e-02 -3.083255e-02
## [1565] -3.173257e-02 -3.282051e-02 -3.305202e-02 -3.358802e-02
## [1569] -3.423512e-02 -3.572438e-02 -3.629587e-02 -3.742851e-02
## [1573] -3.798089e-02 -3.928527e-02 -4.041990e-02 -4.127283e-02
## [1577] -4.191747e-02 -4.247974e-02 -4.315054e-02 -4.362320e-02
## [1581] -4.544339e-02 -4.562467e-02 -4.644750e-02 -4.704507e-02
## [1585] -4.774981e-02 -4.836899e-02 -4.855776e-02 -4.894809e-02
## [1589] -4.930892e-02 -5.159670e-02 -5.262291e-02 -5.350768e-02
## [1593] -5.445369e-02 -5.532854e-02 -5.649021e-02 -5.780863e-02
## [1597] -5.796670e-02 -5.874878e-02 -5.947026e-02 -6.028390e-02
## [1601] -6.075567e-02 -6.141152e-02 -6.211871e-02 -6.337977e-02
## [1605] -6.370947e-02 -6.472525e-02 -6.545775e-02 -6.648843e-02
## [1609] -6.683030e-02 -6.790696e-02 -6.842118e-02 -6.948296e-02
## [1613] -7.072388e-02 -7.077007e-02 -7.214381e-02 -7.401438e-02
## [1617] -7.429393e-02 -7.475699e-02 -7.565848e-02 -7.613555e-02
## [1621] -7.733859e-02 -7.751437e-02 -7.808981e-02 -7.937123e-02
## [1625] -8.049696e-02 -8.137947e-02 -8.171345e-02 -8.254249e-02
## [1629] -8.392501e-02 -8.420250e-02 -8.519134e-02 -8.543039e-02
## [1633] -8.711594e-02 -8.732949e-02 -8.753602e-02 -8.947784e-02
## [1637] -8.966622e-02 -9.009860e-02 -9.074744e-02 -9.172004e-02
## [1641] -9.314211e-02 -9.387763e-02 -9.419092e-02 -9.556312e-02
## [1645] -9.667404e-02 -9.693787e-02 -9.736123e-02 -9.791264e-02
## [1649] -9.944972e-02 -9.952987e-02 -1.015384e-01 -1.019349e-01
## [1653] -1.024735e-01 -1.031263e-01 -1.047205e-01 -1.055582e-01
## [1657] -1.057936e-01 -1.065476e-01 -1.069676e-01 -1.082054e-01
## [1661] -1.085462e-01 -1.097069e-01 -1.104312e-01 -1.118193e-01
## [1665] -1.123152e-01 -1.128147e-01 -1.142206e-01 -1.145056e-01
## [1669] -1.147601e-01 -1.159023e-01 -1.166494e-01 -1.175089e-01
## [1673] -1.182373e-01 -1.186912e-01 -1.189698e-01 -1.208044e-01
## [1677] -1.219271e-01 -1.227802e-01 -1.232563e-01 -1.242693e-01
## [1681] -1.249715e-01 -1.254673e-01 -1.259009e-01 -1.266737e-01
## [1685] -1.268749e-01 -1.281204e-01 -1.292794e-01 -1.299136e-01
## [1689] -1.311100e-01 -1.316783e-01 -1.323067e-01 -1.330200e-01
## [1693] -1.345035e-01 -1.348455e-01 -1.355807e-01 -1.359720e-01
## [1697] -1.370761e-01 -1.376114e-01 -1.390457e-01 -1.393793e-01
## [1701] -1.400964e-01 -1.407575e-01 -1.422022e-01 -1.425205e-01
## [1705] -1.439230e-01 -1.440412e-01 -1.443748e-01 -1.453993e-01
## [1709] -1.468065e-01 -1.475211e-01 -1.480238e-01 -1.486845e-01
## [1713] -1.490770e-01 -1.501799e-01 -1.517138e-01 -1.524676e-01
## [1717] -1.531508e-01 -1.540507e-01 -1.549623e-01 -1.557854e-01
## [1721] -1.561853e-01 -1.571231e-01 -1.577934e-01 -1.580965e-01
## [1725] -1.583558e-01 -1.594044e-01 -1.603051e-01 -1.605288e-01
## [1729] -1.624720e-01 -1.630005e-01 -1.645011e-01 -1.649574e-01
## [1733] -1.661771e-01 -1.666271e-01 -1.667356e-01 -1.674311e-01
## [1737] -1.684744e-01 -1.687854e-01 -1.705487e-01 -1.706426e-01
## [1741] -1.713052e-01 -1.722097e-01 -1.731408e-01 -1.741996e-01
## [1745] -1.749493e-01 -1.754516e-01 -1.758493e-01 -1.766626e-01
## [1749] -1.770392e-01 -1.774679e-01 -1.785130e-01 -1.801248e-01
## [1753] -1.812132e-01 -1.816960e-01 -1.821209e-01 -1.825803e-01
## [1757] -1.839924e-01 -1.846787e-01 -1.848298e-01 -1.860195e-01
## [1761] -1.863928e-01 -1.870720e-01 -1.885345e-01 -1.886855e-01
## [1765] -1.888848e-01 -1.900687e-01 -1.912903e-01 -1.918764e-01
## [1769] -1.933042e-01 -1.936480e-01 -1.950570e-01 -1.960854e-01
## [1773] -1.966140e-01 -1.975293e-01 -1.982352e-01 -1.990229e-01
## [1777] -1.996601e-01 -2.002876e-01 -2.009644e-01 -2.017878e-01
## [1781] -2.024498e-01 -2.029503e-01 -2.032360e-01 -2.047213e-01
## [1785] -2.050666e-01 -2.056971e-01 -2.063476e-01 -2.073152e-01
## [1789] -2.085066e-01 -2.101397e-01 -2.106306e-01 -2.118385e-01
## [1793] -2.123576e-01 -2.125969e-01 -2.133358e-01 -2.138561e-01
## [1797] -2.147336e-01 -2.152546e-01 -2.161518e-01 -2.167686e-01
## [1801] -2.182119e-01 -2.187287e-01 -2.188540e-01 -2.201696e-01
## [1805] -2.212003e-01 -2.215987e-01 -2.222074e-01 -2.230315e-01
## [1809] -2.239366e-01 -2.252825e-01 -2.259659e-01 -2.264439e-01
## [1813] -2.268132e-01 -2.274636e-01 -2.283000e-01 -2.293217e-01
## [1817] -2.302241e-01 -2.309631e-01 -2.314201e-01 -2.330442e-01
## [1821] -2.333843e-01 -2.343750e-01 -2.357779e-01 -2.361988e-01
## [1825] -2.369631e-01 -2.374113e-01 -2.384186e-01 -2.386173e-01
## [1829] -2.388880e-01 -2.397496e-01 -2.409983e-01 -2.414714e-01
## [1833] -2.422519e-01 -2.433564e-01 -2.438041e-01 -2.443840e-01
## [1837] -2.448439e-01 -2.461963e-01 -2.469126e-01 -2.473234e-01
## [1841] -2.478488e-01 -2.490379e-01 -2.502219e-01 -2.511128e-01
## [1845] -2.515687e-01 -2.520231e-01 -2.527000e-01 -2.539644e-01
## [1849] -2.552405e-01 -2.562882e-01 -2.571050e-01 -2.574154e-01
## [1853] -2.591434e-01 -2.593760e-01 -2.605483e-01 -2.607586e-01
## [1857] -2.612510e-01 -2.623397e-01 -2.628138e-01 -2.635598e-01
## [1861] -2.646101e-01 -2.656132e-01 -2.659601e-01 -2.663124e-01
## [1865] -2.677312e-01 -2.677644e-01 -2.689670e-01 -2.695556e-01
## [1869] -2.704014e-01 -2.716002e-01 -2.720225e-01 -2.727114e-01
## [1873] -2.739965e-01 -2.747298e-01 -2.754300e-01 -2.760016e-01
## [1877] -2.766372e-01 -2.770582e-01 -2.780530e-01 -2.782592e-01
## [1881] -2.789966e-01 -2.793964e-01 -2.805097e-01 -2.812598e-01
## [1885] -2.820475e-01 -2.827542e-01 -2.835837e-01 -2.844115e-01
## [1889] -2.853167e-01 -2.857684e-01 -2.872769e-01 -2.880456e-01
## [1893] -2.887668e-01 -2.891568e-01 -2.903054e-01 -2.912273e-01
## [1897] -2.917815e-01 -2.918663e-01 -2.928759e-01 -2.935067e-01
## [1901] -2.944850e-01 -2.952613e-01 -2.960500e-01 -2.967903e-01
## [1905] -2.972110e-01 -2.982693e-01 -2.987268e-01 -2.994339e-01
## [1909] -3.004963e-01 -3.010695e-01 -3.018469e-01 -3.029222e-01
## [1913] -3.038979e-01 -3.043997e-01 -3.060321e-01 -3.076944e-01
## [1917] -3.078806e-01 -3.091613e-01 -3.095886e-01 -3.102634e-01
## [1921] -3.104145e-01 -3.118048e-01 -3.120806e-01 -3.126356e-01
## [1925] -3.132510e-01 -3.142126e-01 -3.151195e-01 -3.159123e-01
## [1929] -3.173174e-01 -3.176247e-01 -3.183998e-01 -3.188336e-01
## [1933] -3.199444e-01 -3.207991e-01 -3.211598e-01 -3.217545e-01
## [1937] -3.225180e-01 -3.237549e-01 -3.242403e-01 -3.256206e-01
## [1941] -3.258592e-01 -3.263629e-01 -3.279132e-01 -3.285490e-01
## [1945] -3.286468e-01 -3.298627e-01 -3.307947e-01 -3.319544e-01
## [1949] -3.326558e-01 -3.333278e-01 -3.335751e-01 -3.343199e-01
## [1953] -3.345676e-01 -3.360647e-01 -3.372735e-01 -3.377613e-01
## [1957] -3.381690e-01 -3.395970e-01 -3.410606e-01 -3.413984e-01
## [1961] -3.419704e-01 -3.433232e-01 -3.438023e-01 -3.445813e-01
## [1965] -3.455479e-01 -3.462429e-01 -3.466129e-01 -3.470961e-01
## [1969] -3.476023e-01 -3.481459e-01 -3.490847e-01 -3.494907e-01
## [1973] -3.512657e-01 -3.520059e-01 -3.524568e-01 -3.530788e-01
## [1977] -3.540316e-01 -3.542684e-01 -3.558897e-01 -3.562244e-01
## [1981] -3.570242e-01 -3.576999e-01 -3.579036e-01 -3.590149e-01
## [1985] -3.605056e-01 -3.609581e-01 -3.617777e-01 -3.624385e-01
## [1989] -3.628114e-01 -3.644651e-01 -3.645640e-01 -3.653860e-01
## [1993] -3.657145e-01 -3.664747e-01 -3.673609e-01 -3.681807e-01
## [1997] -3.684372e-01 -3.691701e-01 -3.699906e-01 -3.710383e-01
## [2001] -3.712910e-01 -3.723091e-01 -3.730085e-01 -3.740178e-01
## [2005] -3.749914e-01 -3.751560e-01 -3.765993e-01 -3.767800e-01
## [2009] -3.780466e-01 -3.785375e-01 -3.797956e-01 -3.806310e-01
## [2013] -3.811759e-01 -3.816665e-01 -3.821151e-01 -3.834564e-01
## [2017] -3.838433e-01 -3.848755e-01 -3.856436e-01 -3.862280e-01
## [2021] -3.873022e-01 -3.880978e-01 -3.890219e-01 -3.897280e-01
## [2025] -3.903626e-01 -3.909279e-01 -3.923421e-01 -3.930570e-01
## [2029] -3.940607e-01 -3.944609e-01 -3.949789e-01 -3.953052e-01
## [2033] -3.965983e-01 -3.969509e-01 -3.981739e-01 -3.987932e-01
## [2037] -3.995885e-01 -4.005164e-01 -4.017165e-01 -4.022076e-01
## [2041] -4.027627e-01 -4.031603e-01 -4.038674e-01 -4.044616e-01
## [2045] -4.056573e-01 -4.062606e-01 -4.067053e-01 -4.070722e-01
## [2049] -4.093734e-01 -4.097644e-01 -4.115672e-01 -4.120845e-01
## [2053] -4.130280e-01 -4.138620e-01 -4.147550e-01 -4.155164e-01
## [2057] -4.156412e-01 -4.167688e-01 -4.177598e-01 -4.185643e-01
## [2061] -4.188417e-01 -4.197213e-01 -4.203551e-01 -4.215160e-01
## [2065] -4.219671e-01 -4.228588e-01 -4.236284e-01 -4.243981e-01
## [2069] -4.259038e-01 -4.268304e-01 -4.275597e-01 -4.281935e-01
## [2073] -4.291643e-01 -4.295545e-01 -4.305679e-01 -4.312809e-01
## [2077] -4.319896e-01 -4.325495e-01 -4.332518e-01 -4.342641e-01
## [2081] -4.351759e-01 -4.366910e-01 -4.369782e-01 -4.379122e-01
## [2085] -4.387744e-01 -4.393820e-01 -4.403994e-01 -4.413486e-01
## [2089] -4.416818e-01 -4.420186e-01 -4.438834e-01 -4.445531e-01
## [2093] -4.455116e-01 -4.455791e-01 -4.461412e-01 -4.467685e-01
## [2097] -4.482707e-01 -4.489459e-01 -4.504225e-01 -4.507802e-01
## [2101] -4.514518e-01 -4.526304e-01 -4.530378e-01 -4.533102e-01
## [2105] -4.536316e-01 -4.552307e-01 -4.560327e-01 -4.564167e-01
## [2109] -4.575859e-01 -4.584715e-01 -4.589243e-01 -4.593881e-01
## [2113] -4.611816e-01 -4.617848e-01 -4.621109e-01 -4.635215e-01
## [2117] -4.637319e-01 -4.649412e-01 -4.660097e-01 -4.666032e-01
## [2121] -4.683996e-01 -4.687113e-01 -4.699086e-01 -4.702954e-01
## [2125] -4.706131e-01 -4.713324e-01 -4.726596e-01 -4.730158e-01
## [2129] -4.739996e-01 -4.745021e-01 -4.747593e-01 -4.753106e-01
## [2133] -4.765290e-01 -4.768966e-01 -4.781446e-01 -4.784540e-01
## [2137] -4.805348e-01 -4.806300e-01 -4.815975e-01 -4.818597e-01
## [2141] -4.824762e-01 -4.835574e-01 -4.844090e-01 -4.850485e-01
## [2145] -4.858033e-01 -4.864587e-01 -4.878550e-01 -4.887003e-01
## [2149] -4.893434e-01 -4.897769e-01 -4.903290e-01 -4.919455e-01
## [2153] -4.927460e-01 -4.933990e-01 -4.937928e-01 -4.944956e-01
## [2157] -4.955127e-01 -4.969867e-01 -4.974580e-01 -4.977362e-01
## [2161] -4.987917e-01 -4.996403e-01 -5.001491e-01 -5.015991e-01
## [2165] -5.026221e-01 -5.026487e-01 -5.030123e-01 -5.038236e-01
## [2169] -5.046579e-01 -5.052221e-01 -5.058748e-01 -5.071322e-01
## [2173] -5.081757e-01 -5.087532e-01 -5.092757e-01 -5.106414e-01
## [2177] -5.116181e-01 -5.130418e-01 -5.136939e-01 -5.139768e-01
## [2181] -5.149230e-01 -5.157644e-01 -5.167315e-01 -5.173506e-01
## [2185] -5.179808e-01 -5.192564e-01 -5.198985e-01 -5.205604e-01
## [2189] -5.210389e-01 -5.219094e-01 -5.237250e-01 -5.244022e-01
## [2193] -5.251531e-01 -5.257431e-01 -5.261199e-01 -5.278237e-01
## [2197] -5.283582e-01 -5.288686e-01 -5.301761e-01 -5.307216e-01
## [2201] -5.314041e-01 -5.321364e-01 -5.332981e-01 -5.337170e-01
## [2205] -5.352044e-01 -5.355551e-01 -5.363639e-01 -5.379508e-01
## [2209] -5.388093e-01 -5.391346e-01 -5.398563e-01 -5.407093e-01
## [2213] -5.422964e-01 -5.426992e-01 -5.438032e-01 -5.441051e-01
## [2217] -5.446635e-01 -5.451943e-01 -5.465968e-01 -5.487743e-01
## [2221] -5.490353e-01 -5.492435e-01 -5.495100e-01 -5.504454e-01
## [2225] -5.514165e-01 -5.517404e-01 -5.525484e-01 -5.535413e-01
## [2229] -5.539108e-01 -5.550694e-01 -5.566398e-01 -5.572176e-01
## [2233] -5.578218e-01 -5.583686e-01 -5.594805e-01 -5.601054e-01
## [2237] -5.608373e-01 -5.616051e-01 -5.631092e-01 -5.644129e-01
## [2241] -5.651926e-01 -5.661513e-01 -5.673524e-01 -5.685688e-01
## [2245] -5.690842e-01 -5.692358e-01 -5.706308e-01 -5.711612e-01
## [2249] -5.719270e-01 -5.723392e-01 -5.735043e-01 -5.752621e-01
## [2253] -5.754210e-01 -5.765653e-01 -5.772080e-01 -5.785331e-01
## [2257] -5.791137e-01 -5.797773e-01 -5.808732e-01 -5.821094e-01
## [2261] -5.824576e-01 -5.839129e-01 -5.851245e-01 -5.856240e-01
## [2265] -5.861768e-01 -5.871592e-01 -5.877132e-01 -5.888483e-01
## [2269] -5.895338e-01 -5.907976e-01 -5.914093e-01 -5.921470e-01
## [2273] -5.930698e-01 -5.936209e-01 -5.951842e-01 -5.959337e-01
## [2277] -5.967312e-01 -5.973795e-01 -5.983511e-01 -5.990959e-01
## [2281] -5.999299e-01 -6.004060e-01 -6.012562e-01 -6.024601e-01
## [2285] -6.041318e-01 -6.050674e-01 -6.057360e-01 -6.062001e-01
## [2289] -6.065177e-01 -6.069434e-01 -6.071043e-01 -6.090072e-01
## [2293] -6.099800e-01 -6.103601e-01 -6.109699e-01 -6.116382e-01
## [2297] -6.131860e-01 -6.141004e-01 -6.152182e-01 -6.159240e-01
## [2301] -6.164654e-01 -6.180394e-01 -6.190416e-01 -6.196658e-01
## [2305] -6.199507e-01 -6.209031e-01 -6.211483e-01 -6.229232e-01
## [2309] -6.236650e-01 -6.241379e-01 -6.247188e-01 -6.257962e-01
## [2313] -6.267502e-01 -6.268131e-01 -6.282034e-01 -6.299524e-01
## [2317] -6.310863e-01 -6.311515e-01 -6.329138e-01 -6.334025e-01
## [2321] -6.338102e-01 -6.345404e-01 -6.360702e-01 -6.365238e-01
## [2325] -6.382337e-01 -6.393031e-01 -6.397021e-01 -6.412019e-01
## [2329] -6.417144e-01 -6.419779e-01 -6.427420e-01 -6.437669e-01
## [2333] -6.449169e-01 -6.458523e-01 -6.460786e-01 -6.473336e-01
## [2337] -6.492735e-01 -6.500850e-01 -6.508303e-01 -6.511631e-01
## [2341] -6.515116e-01 -6.529589e-01 -6.544999e-01 -6.556987e-01
## [2345] -6.571243e-01 -6.577152e-01 -6.592329e-01 -6.597537e-01
## [2349] -6.610358e-01 -6.616503e-01 -6.626409e-01 -6.636514e-01
## [2353] -6.641971e-01 -6.652550e-01 -6.671378e-01 -6.676899e-01
## [2357] -6.692703e-01 -6.694857e-01 -6.706697e-01 -6.724380e-01
## [2361] -6.727405e-01 -6.743215e-01 -6.752100e-01 -6.760221e-01
## [2365] -6.768598e-01 -6.779530e-01 -6.783145e-01 -6.786225e-01
## [2369] -6.796592e-01 -6.816046e-01 -6.828390e-01 -6.849387e-01
## [2373] -6.855589e-01 -6.862747e-01 -6.866211e-01 -6.874805e-01
## [2377] -6.894088e-01 -6.900872e-01 -6.902377e-01 -6.911336e-01
## [2381] -6.913310e-01 -6.925380e-01 -6.929071e-01 -6.937656e-01
## [2385] -6.950447e-01 -6.972695e-01 -6.981215e-01 -6.985693e-01
## [2389] -6.996407e-01 -7.005504e-01 -7.013571e-01 -7.014390e-01
## [2393] -7.024008e-01 -7.041434e-01 -7.059525e-01 -7.066117e-01
## [2397] -7.080113e-01 -7.084214e-01 -7.094348e-01 -7.099873e-01
## [2401] -7.113718e-01 -7.122500e-01 -7.141076e-01 -7.160960e-01
## [2405] -7.164472e-01 -7.180571e-01 -7.184019e-01 -7.190106e-01
## [2409] -7.202969e-01 -7.207608e-01 -7.219155e-01 -7.238476e-01
## [2413] -7.240643e-01 -7.255736e-01 -7.261882e-01 -7.278316e-01
## [2417] -7.295975e-01 -7.303619e-01 -7.319734e-01 -7.327603e-01
## [2421] -7.341933e-01 -7.352122e-01 -7.363202e-01 -7.366143e-01
## [2425] -7.392421e-01 -7.401556e-01 -7.412756e-01 -7.416491e-01
## [2429] -7.435058e-01 -7.452108e-01 -7.457717e-01 -7.472214e-01
## [2433] -7.480059e-01 -7.485601e-01 -7.503011e-01 -7.510209e-01
## [2437] -7.525909e-01 -7.527406e-01 -7.539309e-01 -7.552691e-01
## [2441] -7.560376e-01 -7.569679e-01 -7.585490e-01 -7.591964e-01
## [2445] -7.597083e-01 -7.613917e-01 -7.633293e-01 -7.644921e-01
## [2449] -7.658849e-01 -7.665955e-01 -7.692665e-01 -7.697936e-01
## [2453] -7.712865e-01 -7.719373e-01 -7.734889e-01 -7.743440e-01
## [2457] -7.746455e-01 -7.757996e-01 -7.772636e-01 -7.784809e-01
## [2461] -7.791978e-01 -7.800147e-01 -7.807378e-01 -7.828205e-01
## [2465] -7.835324e-01 -7.853974e-01 -7.859674e-01 -7.877320e-01
## [2469] -7.886073e-01 -7.896368e-01 -7.911370e-01 -7.934708e-01
## [2473] -7.946376e-01 -7.959537e-01 -7.972540e-01 -7.989390e-01
## [2477] -7.995557e-01 -8.001248e-01 -8.012701e-01 -8.043431e-01
## [2481] -8.052798e-01 -8.063297e-01 -8.080776e-01 -8.088622e-01
## [2485] -8.099070e-01 -8.099939e-01 -8.108457e-01 -8.130204e-01
## [2489] -8.144197e-01 -8.164470e-01 -8.173911e-01 -8.178828e-01
## [2493] -8.198154e-01 -8.203534e-01 -8.208488e-01 -8.237349e-01
## [2497] -8.247875e-01 -8.259379e-01 -8.276664e-01 -8.283327e-01
## [2501] -8.294139e-01 -8.316046e-01 -8.328041e-01 -8.346241e-01
## [2505] -8.347898e-01 -8.358497e-01 -8.371983e-01 -8.406719e-01
## [2509] -8.416732e-01 -8.424827e-01 -8.434863e-01 -8.440333e-01
## [2513] -8.468341e-01 -8.488828e-01 -8.508509e-01 -8.520893e-01
## [2517] -8.533449e-01 -8.546998e-01 -8.556689e-01 -8.577624e-01
## [2521] -8.586040e-01 -8.608005e-01 -8.621018e-01 -8.635502e-01
## [2525] -8.643375e-01 -8.664099e-01 -8.674922e-01 -8.694648e-01
## [2529] -8.707802e-01 -8.727218e-01 -8.748411e-01 -8.763657e-01
## [2533] -8.767712e-01 -8.790716e-01 -8.792706e-01 -8.807429e-01
## [2537] -8.811582e-01 -8.825494e-01 -8.863650e-01 -8.871929e-01
## [2541] -8.877796e-01 -8.894906e-01 -8.902449e-01 -8.931704e-01
## [2545] -8.947297e-01 -8.963176e-01 -8.978468e-01 -9.007862e-01
## [2549] -9.010477e-01 -9.023308e-01 -9.039401e-01 -9.049972e-01
## [2553] -9.081756e-01 -9.088427e-01 -9.097581e-01 -9.130090e-01
## [2557] -9.138557e-01 -9.147078e-01 -9.169336e-01 -9.194737e-01
## [2561] -9.215254e-01 -9.256328e-01 -9.275025e-01 -9.292914e-01
## [2565] -9.302889e-01 -9.318631e-01 -9.333906e-01 -9.335973e-01
## [2569] -9.372195e-01 -9.377511e-01 -9.403134e-01 -9.438322e-01
## [2573] -9.444302e-01 -9.471916e-01 -9.490129e-01 -9.493538e-01
## [2577] -9.499172e-01 -9.526532e-01 -9.540852e-01 -9.579994e-01
## [2581] -9.607663e-01 -9.614945e-01 -9.649722e-01 -9.675300e-01
## [2585] -9.692904e-01 -9.710550e-01 -9.730043e-01 -9.767997e-01
## [2589] -9.782272e-01 -9.807540e-01 -9.830686e-01 -9.838844e-01
## [2593] -9.849619e-01 -9.882306e-01 -9.914207e-01 -9.925473e-01
## [2597] -9.961299e-01 -9.978098e-01 -1.001370e+00 -1.002539e+00
## [2601] -1.004439e+00 -1.007577e+00 -1.013514e+00 -1.016435e+00
## [2605] -1.020195e+00 -1.021618e+00 -1.024860e+00 -1.026578e+00
## [2609] -1.030787e+00 -1.033689e+00 -1.034819e+00 -1.035902e+00
## [2613] -1.039176e+00 -1.046242e+00 -1.047559e+00 -1.052570e+00
## [2617] -1.053755e+00 -1.056640e+00 -1.063140e+00 -1.064263e+00
## [2621] -1.067424e+00 -1.069177e+00 -1.072690e+00 -1.076585e+00
## [2625] -1.077383e+00 -1.080870e+00 -1.081678e+00 -1.085491e+00
## [2629] -1.092708e+00 -1.096814e+00 -1.098150e+00 -1.105355e+00
## [2633] -1.107815e+00 -1.108830e+00 -1.114437e+00 -1.121841e+00
## [2637] -1.125104e+00 -1.126824e+00 -1.131623e+00 -1.136020e+00
## [2641] -1.138336e+00 -1.145822e+00 -1.147700e+00 -1.149313e+00
## [2645] -1.156086e+00 -1.159219e+00 -1.164566e+00 -1.169724e+00
## [2649] -1.174185e+00 -1.178681e+00 -1.188530e+00 -1.195898e+00
## [2653] -1.202131e+00 -1.208370e+00 -1.210393e+00 -1.217010e+00
## [2657] -1.225796e+00 -1.237729e+00 -1.244483e+00 -1.251114e+00
## [2661] -1.256676e+00 -1.265003e+00 -1.283626e+00 -1.289836e+00
## [2665] -1.314513e+00 -1.335519e+00 -1.380938e+00
##
## $x
## NULL
##
## $ac
## [1] 0
##
## $GOF
## [1] 0.02803631 0.03628889# MDSplot()
# dim(MDIM$eig)Falta acrescentar Legenda
Plot do modelo rpart - Recursive Partitioning and Regression Trees - personalizando automaticamente a partir do gráfico para o tipo de resposta do modelo.
rp <- rpart::rpart(formula = WR_Grp~.,data=test2)
#rpart::plotcp(rp)
rpart.plot(rp)
rpart.plot.version1(rp)
Falta acrescentar Legenda
5 Conclusão e Considerações
O Random Forest apresenta overfiting para os dados de treinamento, como é possível observar na matriz de confusão obtida, e a partir de 150 arvores de regressão no algoritmo, tem-se um ganho minimo ao adicionar novas árvores.
É possível comparar nossos resultados com os resultados da usuária Yueming (disponíveis no site Kaggle clicando AQUI e AQUI), sendo que a partir de seus resultados é possível obter algumas sugestões de melhora do nosso algoritmo, principalmente quanto ao tratamento dos dados (NA e zeros). Em nosso algoritmo, inicialmente, os filmes repetidos não haviam sido considerados, a autora observou 43 observações com repetição e variáveis com excesso de respostas nulas. Quanto as observações repetidas, replicamos o passo sugerido por ela e foram excluídos todos os filmes com repetição.
Em relação as respostas nulas em excesso, ela tratou essas, como NA, entretanto durante as etapas de preparação de dados, descobrimos que a única variável com tal problema era a “movie_facebook_likes” que acabou sendo excluída de nossa análise.
Ainda foram observadas poucos filmes antes de 1980. Tal situação foi enfrentada pela autora com a exclusãos dessas observações raras. Replicamos esse passo também. De uma forma geral, a autora comparou três modelos (KNN, Árvore de Decisão e Random Forest) apenas levando em consideração a acurácia obtida nos modelos.
Aqui, tentamos levar em consideração os valores, principalmente, da matriz de confusão e não somente a acurácia. E podemos concluir que:
O nosso modelo é muito bom para classificar filmes com alta e baixa aprovação. Utilizando a base de teste, o modelo retornou através da matriz de confusão uma sensibilidade de cerca de 85% e uma especificidade de 80%. Além disso, uma acurácia obtida de 83%.
É valido ressaltar, também, que o erro OOB estimado foi de 20%.
Ainda através da matriz de confusão, o nosso modelo erra em uma menor quantidade de vezes uma classificação de filmes de até 6,5★ (baixa aprovação). A estimativa de erro de classificação desses filmes é de 18% versus 22% de erro de classificação em filmes com alta aprovação.
O quadro a seguir apresenta o número de filmes em cada classificação segundo a base de teste, treino e inteira (movie). Aparentemente as categorias são bem balanceadas.
| Base de dados | [0, 6.5) | [6.5, 10] |
|---|---|---|
| Movie (treino + teste) | 1883 | 1883 |
| Treino | 1328 | 1339 |
| Teste | 555 | 544 |
Ainda conseguimos obter quais são os fatores/variáveis mais importantes para análise de classificação, dentre as variáveis existentes. Abaixo citamos as top 10 variáveis mais importantes.
RF1 = RF1[1:10,]
hc6_1 <- highchart() %>%
hc_add_series(data = RF1$importance,
type = "bar",
name = "Importância",
showInLegend = FALSE,
tooltip = list(valueDecimals = 2, valuePrefix = "", valueSuffix = ""),
color="orange") %>%
hc_yAxis(title = list(text = "Importância"),
allowDecimals = TRUE, max = 200,
labels = list(format = "{value}")) %>%
hc_xAxis(title = list(text = "Fatores"),
categories = RF1$variables,
tickmarkPlacement = "on",
opposite = FALSE) %>%
hc_title(text = "Importância por fator - Random Forest",
style = list(fontWeight = "bold")) %>%
hc_subtitle(text = paste("")) %>%
hc_tooltip(valueDecimals = 2,
pointFormat = "Importância: {point.y}")%>%
#pointFormat = "Variável: {point.x} <br> Importância: {point.y}")
hc_credits(enabled = TRUE,
text = "Fonte: IMDB/KAGGLE. Elaboração: Ewerson Pimenta.",
style = list(fontSize = "10px")) %>%
hc_exporting(enabled = TRUE, filename = "F6_1-importance-Pimenta")
#hc <- hc %>%
# hc_add_theme(hc_theme_darkunica())
hc6_1Tempo de execução
proc.time() -ptm## user system elapsed
## 59.26 1.95 66.64